| amazon-nova-experimental-chat (26-02-10) | Arena text: business and management | 1,450 | amazon-nova-experimental-chat (26-02-10) | 26-02-10 | Arena score | CC-BY-4.0 | lmarena |
|---|
| amazon-nova-experimental-chat (26-02-10) | Arena text: coding | 1,483 | amazon-nova-experimental-chat (26-02-10) | 26-02-10 | Arena score | CC-BY-4.0 | lmarena |
|---|
| amazon-nova-experimental-chat (26-02-10) | Arena text: creative writing | 1,369 | amazon-nova-experimental-chat (26-02-10) | 26-02-10 | Arena score | CC-BY-4.0 | lmarena |
|---|
| amazon-nova-experimental-chat (26-02-10) | Arena text: following instructions | 1,427 | amazon-nova-experimental-chat (26-02-10) | 26-02-10 | Arena score | CC-BY-4.0 | lmarena |
|---|
| amazon-nova-experimental-chat (26-02-10) | Arena text: maths | 1,437 | amazon-nova-experimental-chat (26-02-10) | 26-02-10 | Arena score | CC-BY-4.0 | lmarena |
|---|
| amazon-nova-experimental-chat (26-02-10) | Arena text: overall preference | 1,449 | amazon-nova-experimental-chat (26-02-10) | 26-02-10 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5 | Arena text: business and management | 1,480 | Claude Fable 5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5 | Arena text: coding | 1,518 | Claude Fable 5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5 | Arena text: creative writing | 1,493 | Claude Fable 5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5 | Arena text: following instructions | 1,503 | Claude Fable 5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5 | Arena text: maths | 1,516 | Claude Fable 5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5 | Arena text: overall preference | 1,492 | Claude Fable 5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5 | Arena WebDev: building web apps | 1,624 | Claude Fable 5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5 | OTIS Mock AIME 2024-2025 | 99.7% | claude-fable-5_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Fable 5 | FrontierCode (reported by Cognition) | 53.5% | claude-fable-5_unknown | xhigh, claude-code | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Claude Fable 5 | FrontierMath Tiers 1-3 (v2) | 87.0% | claude-fable-5_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Fable 5 | GPQA Diamond | 85.9% | claude-fable-5_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Fable 5 | SimpleQA Verified | 70.7% | claude-fable-5_xhigh | xhigh | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Fable 5.1 | Arena text: business and management | 1,494 | Claude Fable 5.1 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5.1 | Arena text: coding | 1,530 | Claude Fable 5.1 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5.1 | Arena text: creative writing | 1,512 | Claude Fable 5.1 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5.1 | Arena text: following instructions | 1,519 | Claude Fable 5.1 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5.1 | Arena text: maths | 1,531 | Claude Fable 5.1 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5.1 | Arena text: overall preference | 1,514 | Claude Fable 5.1 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5.1 | Arena WebDev: building web apps | 1,746 | Claude Fable 5.1 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Fable 5.1 | OTIS Mock AIME 2024-2025 | 100.0% | claude-fable-5-1_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Fable 5.1 | FrontierCode (reported by Cognition) | 50.9% | claude-fable-5-1_medium | medium, claude-code | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Claude Fable 5.1 | FrontierMath Tiers 1-3 (v2) | 90.2% | claude-fable-5-1_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Fable 5.1 | SimpleQA Verified | 70.8% | claude-fable-5-1_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Haiku 5.5 | Arena text: business and management | 1,405 | Claude Haiku 5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Haiku 5.5 | Arena text: coding | 1,454 | Claude Haiku 5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Haiku 5.5 | Arena text: creative writing | 1,336 | Claude Haiku 5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Haiku 5.5 | Arena text: following instructions | 1,415 | Claude Haiku 5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Haiku 5.5 | Arena text: maths | 1,434 | Claude Haiku 5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Haiku 5.5 | Arena text: overall preference | 1,401 | Claude Haiku 5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Haiku 5.5 | Arena WebDev: building web apps | 1,589 | Claude Haiku 5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Haiku 5.5 | OTIS Mock AIME 2024-2025 | 98.9% | claude-haiku-5-5_xhigh | xhigh | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Haiku 5.5 | FrontierCode (reported by Cognition) | 46.4% | claude-haiku-5-5_max | max, claude-code | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Claude Haiku 5.5 | FrontierMath Tiers 1-3 (v2) | 75.1% | claude-haiku-5-5_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Haiku 5.5 | GPQA Diamond | 89.6% | claude-haiku-5-5_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Haiku 5.5 | SimpleQA Verified | 23.8% | claude-haiku-5-5_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.5 (20251101) | Arena text: business and management | 1,448 | Claude Opus 4.5 (20251101) | 20251101 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.5 (20251101) | Arena text: coding | 1,498 | Claude Opus 4.5 (20251101) | 20251101 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.5 (20251101) | Arena text: creative writing | 1,443 | Claude Opus 4.5 (20251101) | 20251101 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.5 (20251101) | Arena text: following instructions | 1,472 | Claude Opus 4.5 (20251101) | 20251101 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.5 (20251101) | Arena text: maths | 1,458 | Claude Opus 4.5 (20251101) | 20251101 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.5 (20251101) | Arena text: overall preference | 1,451 | Claude Opus 4.5 (20251101) | 20251101 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.5 (20251101) | Arena WebDev: building web apps | 1,469 | Claude Opus 4.5 (20251101) | 20251101 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.6 | Arena text: business and management | 1,497 | Claude Opus 4.6 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.6 | Arena text: coding | 1,531 | Claude Opus 4.6 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.6 | Arena text: creative writing | 1,504 | Claude Opus 4.6 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.6 | Arena text: following instructions | 1,521 | Claude Opus 4.6 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.6 | Arena text: maths | 1,518 | Claude Opus 4.6 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.6 | Arena text: overall preference | 1,502 | Claude Opus 4.6 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.6 | Arena WebDev: building web apps | 1,546 | Claude Opus 4.6 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.6 | OTIS Mock AIME 2024-2025 | 91.1% | claude-opus-4-6_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.6 | FrontierCode (reported by Cognition) | 26.6% | claude-opus-4-6_unknown | high, claude-code | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.6 | FrontierMath Tiers 1-3 (v2) | 66.0% | claude-opus-4-6_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.6 | GPQA Diamond | 88.4% | claude-opus-4-6_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.6 | SimpleQA Verified | 47.0% | claude-opus-4-6_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.7 | Arena text: business and management | 1,490 | Claude Opus 4.7 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.7 | Arena text: coding | 1,519 | Claude Opus 4.7 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.7 | Arena text: creative writing | 1,486 | Claude Opus 4.7 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.7 | Arena text: following instructions | 1,498 | Claude Opus 4.7 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.7 | Arena text: maths | 1,499 | Claude Opus 4.7 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.7 | Arena text: overall preference | 1,490 | Claude Opus 4.7 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.7 | Arena WebDev: building web apps | 1,558 | Claude Opus 4.7 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.7 | OTIS Mock AIME 2024-2025 | 86.7% | claude-opus-4-7_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.7 | FrontierCode (reported by Cognition) | 38.5% | claude-opus-4-7_unknown | max, claude-code | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.7 | FrontierMath Tiers 1-3 (v2) | 70.2% | claude-opus-4-7_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.7 | GPQA Diamond | 86.4% | claude-opus-4-7_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.7 | SimpleQA Verified | 51.7% | claude-opus-4-7_xhigh | xhigh | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.7 | SWE-bench Verified (Epoch runs) | 83.5% | claude-opus-4-7_max | max | share of issues resolved | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.8 | Arena text: business and management | 1,458 | Claude Opus 4.8 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.8 | Arena text: coding | 1,491 | Claude Opus 4.8 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.8 | Arena text: creative writing | 1,455 | Claude Opus 4.8 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.8 | Arena text: following instructions | 1,477 | Claude Opus 4.8 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.8 | Arena text: maths | 1,485 | Claude Opus 4.8 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.8 | Arena text: overall preference | 1,461 | Claude Opus 4.8 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.8 | Arena WebDev: building web apps | 1,556 | Claude Opus 4.8 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 4.8 | OTIS Mock AIME 2024-2025 | 98.3% | claude-opus-4-8_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.8 | FrontierCode (reported by Cognition) | 46.5% | claude-opus-4-8_unknown | max, claude-code | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.8 | FrontierMath Tiers 1-3 (v2) | 80.0% | claude-opus-4-8_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.8 | GPQA Diamond | 91.0% | claude-opus-4-8_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Opus 4.8 | SimpleQA Verified | 53.0% | claude-opus-4-8_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Opus 5 | Arena text: business and management | 1,500 | Claude Opus 5 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 5 | Arena text: coding | 1,532 | Claude Opus 5 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 5 | Arena text: creative writing | 1,495 | Claude Opus 5 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 5 | Arena text: following instructions | 1,518 | Claude Opus 5 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 5 | Arena text: maths | 1,531 | Claude Opus 5 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 5 | Arena text: overall preference | 1,508 | Claude Opus 5 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 5 | Arena WebDev: building web apps | 1,691 | Claude Opus 5 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 5 | OTIS Mock AIME 2024-2025 | 98.9% | claude-opus-5_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Opus 5 | FrontierCode (reported by Cognition) | 53.4% | claude-opus-5_max | medium, claude-code | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Claude Opus 5 | FrontierMath Tiers 1-3 (v2) | 85.6% | claude-opus-5_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Opus 5 | GPQA Diamond | 93.9% | claude-opus-5_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Opus 5 | SimpleQA Verified | 59.9% | claude-opus-5_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Opus 5.5 | Arena text: business and management | 1,497 | Claude Opus 5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 5.5 | Arena text: coding | 1,547 | Claude Opus 5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 5.5 | Arena text: creative writing | 1,534 | Claude Opus 5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 5.5 | Arena text: following instructions | 1,539 | Claude Opus 5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 5.5 | Arena text: maths | 1,517 | Claude Opus 5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 5.5 | Arena text: overall preference | 1,516 | Claude Opus 5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 5.5 | Arena WebDev: building web apps | 1,810 | Claude Opus 5.5 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Opus 5.5 | OTIS Mock AIME 2024-2025 | 100.0% | claude-opus-5-5_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Opus 5.5 | FrontierCode (reported by Cognition) | 54.6% | claude-opus-5-5_medium | medium, claude-code | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Claude Opus 5.5 | FrontierMath Tiers 1-3 (v2) | 91.2% | claude-opus-5-5_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Opus 5.5 | GPQA Diamond | 90.6% | claude-opus-5-5_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Opus 5.5 | SimpleQA Verified | 72.2% | claude-opus-5-5_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 4.6 | Arena text: business and management | 1,464 | Claude Sonnet 4.6 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 4.6 | Arena text: coding | 1,503 | Claude Sonnet 4.6 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 4.6 | Arena text: creative writing | 1,435 | Claude Sonnet 4.6 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 4.6 | Arena text: following instructions | 1,476 | Claude Sonnet 4.6 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 4.6 | Arena text: maths | 1,462 | Claude Sonnet 4.6 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 4.6 | Arena text: overall preference | 1,458 | Claude Sonnet 4.6 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 4.6 | Arena WebDev: building web apps | 1,523 | Claude Sonnet 4.6 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 4.6 | OTIS Mock AIME 2024-2025 | 71.1% | claude-sonnet-4-6_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 4.6 | FrontierCode (reported by Cognition) | 24.3% | claude-sonnet-4-6_unknown | max, claude-code | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 4.6 | GPQA Diamond | 78.8% | claude-sonnet-4-6_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 4.6 | SimpleQA Verified | 32.8% | claude-sonnet-4-6_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 4.6 | SWE-bench Verified (Epoch runs) | 75.2% | claude-sonnet-4-6 | not stated | share of issues resolved | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 5 | Arena text: business and management | 1,437 | Claude Sonnet 5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 5 | Arena text: coding | 1,484 | Claude Sonnet 5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 5 | Arena text: creative writing | 1,416 | Claude Sonnet 5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 5 | Arena text: following instructions | 1,453 | Claude Sonnet 5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 5 | Arena text: maths | 1,469 | Claude Sonnet 5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 5 | Arena text: overall preference | 1,443 | Claude Sonnet 5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 5 | Arena WebDev: building web apps | 1,540 | Claude Sonnet 5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 5 | OTIS Mock AIME 2024-2025 | 80.0% | claude-sonnet-5_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 5 | FrontierCode (reported by Cognition) | 42.7% | claude-sonnet-5_unknown | xhigh, claude-code | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 5 | FrontierMath Tiers 1-3 (v2) | 65.6% | claude-sonnet-5_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 5 | GPQA Diamond | 80.3% | claude-sonnet-5_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 5 | SimpleQA Verified | 33.7% | claude-sonnet-5_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 5.5 | Arena text: business and management | 1,466 | Claude Sonnet 5.5 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 5.5 | Arena text: coding | 1,512 | Claude Sonnet 5.5 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 5.5 | Arena text: creative writing | 1,470 | Claude Sonnet 5.5 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 5.5 | Arena text: following instructions | 1,492 | Claude Sonnet 5.5 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 5.5 | Arena text: maths | 1,504 | Claude Sonnet 5.5 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 5.5 | Arena text: overall preference | 1,473 | Claude Sonnet 5.5 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 5.5 | Arena WebDev: building web apps | 1,770 | Claude Sonnet 5.5 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Claude Sonnet 5.5 | OTIS Mock AIME 2024-2025 | 100.0% | claude-sonnet-5-5_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 5.5 | FrontierCode (reported by Cognition) | 52.1% | claude-sonnet-5-5_xhigh | xhigh, claude-code | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 5.5 | FrontierMath Tiers 1-3 (v2) | 88.8% | claude-sonnet-5-5_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 5.5 | GPQA Diamond | 95.6% | claude-sonnet-5-5_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Claude Sonnet 5.5 | SimpleQA Verified | 46.5% | claude-sonnet-5-5_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| DeepSeek-V4.1-Flash-20260910 | Arena text: business and management | 1,462 | DeepSeek-V4.1-Flash-20260910 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4.1-Flash-20260910 | Arena text: coding | 1,501 | DeepSeek-V4.1-Flash-20260910 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4.1-Flash-20260910 | Arena text: creative writing | 1,437 | DeepSeek-V4.1-Flash-20260910 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4.1-Flash-20260910 | Arena text: following instructions | 1,474 | DeepSeek-V4.1-Flash-20260910 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4.1-Flash-20260910 | Arena text: maths | 1,481 | DeepSeek-V4.1-Flash-20260910 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4.1-Flash-20260910 | Arena text: overall preference | 1,465 | DeepSeek-V4.1-Flash-20260910 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4.1-Flash-20260910 | Arena WebDev: building web apps | 1,619 | DeepSeek-V4.1-Flash-20260910 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro | Arena text: business and management | 1,443 | DeepSeek-V4-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro | Arena text: coding | 1,469 | DeepSeek-V4-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro | Arena text: creative writing | 1,446 | DeepSeek-V4-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro | Arena text: following instructions | 1,445 | DeepSeek-V4-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro | Arena text: maths | 1,442 | DeepSeek-V4-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro | Arena text: overall preference | 1,451 | DeepSeek-V4-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro | Arena WebDev: building web apps | 1,447 | DeepSeek-V4-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro | OTIS Mock AIME 2024-2025 | 96.7% | deepseek-v4-pro_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| DeepSeek-V4-Pro | FrontierCode (reported by Cognition) | 17.6% | deepseek-v4-pro_none | none, mini-swe-agent | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| DeepSeek-V4-Pro | FrontierMath Tiers 1-3 (v2) | 45.3% | deepseek-v4-pro_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| DeepSeek-V4-Pro | GPQA Diamond | 89.6% | deepseek-v4-pro_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| DeepSeek-V4-Pro | SimpleQA Verified | 47.0% | deepseek-v4-pro_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| DeepSeek-V4-Pro | SWE-bench Verified (Epoch runs) | 77.6% | deepseek-v4-pro_max | max | share of issues resolved | CC-BY-4.0 | epoch |
|---|
| DeepSeek-V4-Pro (High, 20260813) | Arena text: business and management | 1,442 | DeepSeek-V4-Pro (High, 20260813) | High, 20260813 | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro (High, 20260813) | Arena text: coding | 1,469 | DeepSeek-V4-Pro (High, 20260813) | High, 20260813 | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro (High, 20260813) | Arena text: creative writing | 1,440 | DeepSeek-V4-Pro (High, 20260813) | High, 20260813 | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro (High, 20260813) | Arena text: following instructions | 1,449 | DeepSeek-V4-Pro (High, 20260813) | High, 20260813 | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro (High, 20260813) | Arena text: maths | 1,442 | DeepSeek-V4-Pro (High, 20260813) | High, 20260813 | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro (High, 20260813) | Arena text: overall preference | 1,446 | DeepSeek-V4-Pro (High, 20260813) | High, 20260813 | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro (High, 20260813) | Arena WebDev: building web apps | 1,582 | DeepSeek-V4-Pro (High, 20260813) | High, 20260813 | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro (High, 20260813) | OTIS Mock AIME 2024-2025 | 98.6% | deepseek-v4-pro-0813_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| DeepSeek-V4-Pro (High, 20260813) | FrontierCode (reported by Cognition) | 28.5% | deepseek-v4-pro-0813_high | high, devin-cli | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| DeepSeek-V4-Pro (High, 20260813) | FrontierMath Tiers 1-3 (v2) | 64.6% | deepseek-v4-pro-0813_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| DeepSeek-V4-Pro (High, 20260813) | GPQA Diamond | 91.7% | deepseek-v4-pro-0813_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| DeepSeek-V4-Pro (High, 20260813) | SimpleQA Verified | 52.9% | deepseek-v4-pro-0813_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| DeepSeek-V4-Pro Preview | Arena text: business and management | 1,428 | DeepSeek-V4-Pro Preview (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro Preview | Arena text: coding | 1,453 | DeepSeek-V4-Pro Preview (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro Preview | Arena text: creative writing | 1,441 | DeepSeek-V4-Pro Preview (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro Preview | Arena text: following instructions | 1,436 | DeepSeek-V4-Pro Preview (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro Preview | Arena text: maths | 1,454 | DeepSeek-V4-Pro Preview (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro Preview | Arena text: overall preference | 1,445 | DeepSeek-V4-Pro Preview (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| DeepSeek-V4-Pro Preview | Arena WebDev: building web apps | 1,464 | DeepSeek-V4-Pro Preview (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Dola Seed 2.0 Pro | Arena text: business and management | 1,439 | Dola Seed 2.0 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Dola Seed 2.0 Pro | Arena text: coding | 1,472 | Dola Seed 2.0 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Dola Seed 2.0 Pro | Arena text: creative writing | 1,406 | Dola Seed 2.0 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Dola Seed 2.0 Pro | Arena text: following instructions | 1,414 | Dola Seed 2.0 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Dola Seed 2.0 Pro | Arena text: maths | 1,439 | Dola Seed 2.0 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Dola Seed 2.0 Pro | Arena text: overall preference | 1,448 | Dola Seed 2.0 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| ERNIE 5.0 (0110) | Arena text: business and management | 1,429 | ERNIE 5.0 (0110) | 0110 | Arena score | CC-BY-4.0 | lmarena |
|---|
| ERNIE 5.0 (0110) | Arena text: coding | 1,455 | ERNIE 5.0 (0110) | 0110 | Arena score | CC-BY-4.0 | lmarena |
|---|
| ERNIE 5.0 (0110) | Arena text: creative writing | 1,426 | ERNIE 5.0 (0110) | 0110 | Arena score | CC-BY-4.0 | lmarena |
|---|
| ERNIE 5.0 (0110) | Arena text: following instructions | 1,413 | ERNIE 5.0 (0110) | 0110 | Arena score | CC-BY-4.0 | lmarena |
|---|
| ERNIE 5.0 (0110) | Arena text: maths | 1,436 | ERNIE 5.0 (0110) | 0110 | Arena score | CC-BY-4.0 | lmarena |
|---|
| ERNIE 5.0 (0110) | Arena text: overall preference | 1,445 | ERNIE 5.0 (0110) | 0110 | Arena score | CC-BY-4.0 | lmarena |
|---|
| ERNIE 5.1 | Arena text: business and management | 1,473 | ERNIE 5.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| ERNIE 5.1 | Arena text: coding | 1,488 | ERNIE 5.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| ERNIE 5.1 | Arena text: creative writing | 1,441 | ERNIE 5.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| ERNIE 5.1 | Arena text: following instructions | 1,460 | ERNIE 5.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| ERNIE 5.1 | Arena text: maths | 1,482 | ERNIE 5.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| ERNIE 5.1 | Arena text: overall preference | 1,468 | ERNIE 5.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 2.5 Pro | Arena text: business and management | 1,439 | Gemini 2.5 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 2.5 Pro | Arena text: coding | 1,451 | Gemini 2.5 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 2.5 Pro | Arena text: creative writing | 1,454 | Gemini 2.5 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 2.5 Pro | Arena text: following instructions | 1,437 | Gemini 2.5 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 2.5 Pro | Arena text: maths | 1,450 | Gemini 2.5 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 2.5 Pro | Arena text: overall preference | 1,458 | Gemini 2.5 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 2.5 Pro | Arena WebDev: building web apps | 1,226 | Gemini 2.5 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.1 Pro Preview | Arena text: business and management | 1,459 | Gemini 3.1 Pro Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.1 Pro Preview | Arena text: coding | 1,484 | Gemini 3.1 Pro Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.1 Pro Preview | Arena text: creative writing | 1,482 | Gemini 3.1 Pro Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.1 Pro Preview | Arena text: following instructions | 1,466 | Gemini 3.1 Pro Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.1 Pro Preview | Arena text: maths | 1,485 | Gemini 3.1 Pro Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.1 Pro Preview | Arena text: overall preference | 1,480 | Gemini 3.1 Pro Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.1 Pro Preview | Arena WebDev: building web apps | 1,447 | Gemini 3.1 Pro Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash | Arena text: business and management | 1,467 | Gemini 3.5 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash | Arena text: coding | 1,493 | Gemini 3.5 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash | Arena text: creative writing | 1,469 | Gemini 3.5 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash | Arena text: following instructions | 1,467 | Gemini 3.5 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash | Arena text: maths | 1,504 | Gemini 3.5 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash | Arena text: overall preference | 1,482 | Gemini 3.5 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash | Arena WebDev: building web apps | 1,499 | Gemini 3.5 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash | OTIS Mock AIME 2024-2025 | 95.6% | gemini-3.5-flash_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Gemini 3.5 Flash | FrontierMath Tiers 1-3 (v2) | 62.8% | gemini-3.5-flash_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Gemini 3.5 Flash | GPQA Diamond | 92.8% | gemini-3.5-flash_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Gemini 3.5 Flash | SimpleQA Verified | 66.2% | gemini-3.5-flash_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Gemini 3.5 Flash | SWE-bench Verified (Epoch runs) | 79.3% | gemini-3.5-flash_high | high | share of issues resolved | CC-BY-4.0 | epoch |
|---|
| Gemini 3.5 Flash-Lite | Arena text: business and management | 1,423 | Gemini 3.5 Flash-Lite | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash-Lite | Arena text: coding | 1,453 | Gemini 3.5 Flash-Lite | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash-Lite | Arena text: creative writing | 1,421 | Gemini 3.5 Flash-Lite | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash-Lite | Arena text: following instructions | 1,420 | Gemini 3.5 Flash-Lite | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash-Lite | Arena text: maths | 1,436 | Gemini 3.5 Flash-Lite | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash-Lite | Arena text: overall preference | 1,435 | Gemini 3.5 Flash-Lite | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash-Lite | Arena WebDev: building web apps | 1,434 | Gemini 3.5 Flash-Lite | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.5 Flash-Lite | OTIS Mock AIME 2024-2025 | 71.1% | gemini-3.5-flash-lite_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Gemini 3.5 Flash-Lite | FrontierMath Tiers 1-3 (v2) | 26.0% | gemini-3.5-flash-lite_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Gemini 3.5 Flash-Lite | GPQA Diamond | 83.3% | gemini-3.5-flash-lite_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Gemini 3.6 Flash | Arena text: business and management | 1,464 | Gemini 3.6 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.6 Flash | Arena text: coding | 1,491 | Gemini 3.6 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.6 Flash | Arena text: creative writing | 1,466 | Gemini 3.6 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.6 Flash | Arena text: following instructions | 1,467 | Gemini 3.6 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.6 Flash | Arena text: maths | 1,507 | Gemini 3.6 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.6 Flash | Arena text: overall preference | 1,480 | Gemini 3.6 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.6 Flash | Arena WebDev: building web apps | 1,538 | Gemini 3.6 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.6 Flash | OTIS Mock AIME 2024-2025 | 94.2% | gemini-3.6-flash_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Gemini 3.6 Flash | FrontierCode (reported by Cognition) | 34.4% | gemini-3.6-flash_unknown | medium, devin-cli | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Gemini 3.6 Flash | FrontierMath Tiers 1-3 (v2) | 58.9% | gemini-3.6-flash_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Gemini 3.6 Flash | GPQA Diamond | 94.1% | gemini-3.6-flash_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Gemini 3.6 Flash | SimpleQA Verified | 66.2% | gemini-3.6-flash_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Gemini 3.7 Flash | Arena text: business and management | 1,480 | Gemini 3.7 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.7 Flash | Arena text: coding | 1,500 | Gemini 3.7 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.7 Flash | Arena text: creative writing | 1,487 | Gemini 3.7 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.7 Flash | Arena text: following instructions | 1,486 | Gemini 3.7 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.7 Flash | Arena text: maths | 1,507 | Gemini 3.7 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.7 Flash | Arena text: overall preference | 1,489 | Gemini 3.7 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.7 Flash | Arena WebDev: building web apps | 1,590 | Gemini 3.7 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.7 Flash | OTIS Mock AIME 2024-2025 | 97.2% | gemini-3.7-flash_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Gemini 3.7 Flash | FrontierCode (reported by Cognition) | 43.6% | gemini-3.7-flash_unknown | medium, devin-cli | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Gemini 3.7 Flash | FrontierMath Tiers 1-3 (v2) | 71.6% | gemini-3.7-flash_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Gemini 3.7 Flash | GPQA Diamond | 94.8% | gemini-3.7-flash_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Gemini 3.7 Flash | SimpleQA Verified | 69.2% | gemini-3.7-flash_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Gemini 3.8 Flash | Arena text: business and management | 1,480 | Gemini 3.8 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.8 Flash | Arena text: coding | 1,509 | Gemini 3.8 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.8 Flash | Arena text: creative writing | 1,491 | Gemini 3.8 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.8 Flash | Arena text: following instructions | 1,493 | Gemini 3.8 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.8 Flash | Arena text: maths | 1,530 | Gemini 3.8 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.8 Flash | Arena text: overall preference | 1,498 | Gemini 3.8 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.8 Flash | Arena WebDev: building web apps | 1,583 | Gemini 3.8 Flash (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3.8 Flash | OTIS Mock AIME 2024-2025 | 98.9% | gemini-3.8-flash_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Gemini 3.8 Flash | FrontierCode (reported by Cognition) | 41.2% | gemini-3.8-flash_medium | medium, devin-cli | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Gemini 3.8 Flash | FrontierMath Tiers 1-3 (v2) | 68.4% | gemini-3.8-flash_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Gemini 3.8 Flash | GPQA Diamond | 95.4% | gemini-3.8-flash_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Gemini 3.8 Flash | SimpleQA Verified | 69.7% | gemini-3.8-flash_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Gemini 3 Flash | Arena text: business and management | 1,447 | Gemini 3 Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3 Flash | Arena text: coding | 1,459 | Gemini 3 Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3 Flash | Arena text: creative writing | 1,457 | Gemini 3 Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3 Flash | Arena text: following instructions | 1,436 | Gemini 3 Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3 Flash | Arena text: maths | 1,473 | Gemini 3 Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3 Flash | Arena text: overall preference | 1,466 | Gemini 3 Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3 Flash | Arena WebDev: building web apps | 1,440 | Gemini 3 Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3 Pro | Arena text: business and management | 1,458 | Gemini 3 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3 Pro | Arena text: coding | 1,481 | Gemini 3 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3 Pro | Arena text: creative writing | 1,482 | Gemini 3 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3 Pro | Arena text: following instructions | 1,458 | Gemini 3 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3 Pro | Arena text: maths | 1,476 | Gemini 3 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3 Pro | Arena text: overall preference | 1,479 | Gemini 3 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 3 Pro | Arena WebDev: building web apps | 1,440 | Gemini 3 Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 4 Argon | Arena text: business and management | 1,515 | Gemini 4 Argon (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 4 Argon | Arena text: coding | 1,538 | Gemini 4 Argon (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 4 Argon | Arena text: creative writing | 1,527 | Gemini 4 Argon (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 4 Argon | Arena text: following instructions | 1,533 | Gemini 4 Argon (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 4 Argon | Arena text: maths | 1,538 | Gemini 4 Argon (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 4 Argon | Arena text: overall preference | 1,529 | Gemini 4 Argon (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 4 Argon | Arena WebDev: building web apps | 1,679 | Gemini 4 Argon (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemini 4 Argon | OTIS Mock AIME 2024-2025 | 100.0% | gemini-4-argon_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Gemini 4 Argon | FrontierCode (reported by Cognition) | 47.5% | gemini-4-argon_medium | medium, devin-cli | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Gemini 4 Argon | FrontierMath Tiers 1-3 (v2) | 86.0% | gemini-4-argon_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Gemini 4 Argon | GPQA Diamond | 96.4% | gemini-4-argon_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Gemini 4 Argon | SimpleQA Verified | 76.0% | gemini-4-argon_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Gemma 4 31B | Arena text: business and management | 1,430 | Gemma 4 31B | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemma 4 31B | Arena text: coding | 1,459 | Gemma 4 31B | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemma 4 31B | Arena text: creative writing | 1,416 | Gemma 4 31B | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemma 4 31B | Arena text: following instructions | 1,434 | Gemma 4 31B | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemma 4 31B | Arena text: maths | 1,464 | Gemma 4 31B | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemma 4 31B | Arena text: overall preference | 1,443 | Gemma 4 31B | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Gemma 4 31B | Arena WebDev: building web apps | 1,369 | Gemma 4 31B | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-4.7 | Arena text: business and management | 1,429 | GLM-4.7 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-4.7 | Arena text: coding | 1,454 | GLM-4.7 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-4.7 | Arena text: creative writing | 1,401 | GLM-4.7 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-4.7 | Arena text: following instructions | 1,412 | GLM-4.7 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-4.7 | Arena text: maths | 1,424 | GLM-4.7 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-4.7 | Arena text: overall preference | 1,435 | GLM-4.7 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-4.7 | Arena WebDev: building web apps | 1,435 | GLM-4.7 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-4.7 | OTIS Mock AIME 2024-2025 | 83.3% | glm-4.7 | not stated | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GLM-4.7 | GPQA Diamond | 83.3% | glm-4.7 | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GLM-4.7 | SimpleQA Verified | 32.2% | glm-4.7 | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GLM-5 | Arena text: business and management | 1,430 | GLM-5 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5 | Arena text: coding | 1,461 | GLM-5 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5 | Arena text: creative writing | 1,438 | GLM-5 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5 | Arena text: following instructions | 1,428 | GLM-5 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5 | Arena text: maths | 1,440 | GLM-5 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5 | Arena text: overall preference | 1,446 | GLM-5 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5 | Arena WebDev: building web apps | 1,436 | GLM-5 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5 | OTIS Mock AIME 2024-2025 | 80.0% | glm-5 | not stated | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GLM-5 | GPQA Diamond | 87.8% | glm-5 | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GLM-5 | SWE-bench Verified (Epoch runs) | 72.1% | glm-5 | not stated | share of issues resolved | CC-BY-4.0 | epoch |
|---|
| GLM-5.1 | Arena text: business and management | 1,453 | GLM-5.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.1 | Arena text: coding | 1,486 | GLM-5.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.1 | Arena text: creative writing | 1,453 | GLM-5.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.1 | Arena text: following instructions | 1,452 | GLM-5.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.1 | Arena text: maths | 1,473 | GLM-5.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.1 | Arena text: overall preference | 1,463 | GLM-5.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.1 | Arena WebDev: building web apps | 1,508 | GLM-5.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.1 | OTIS Mock AIME 2024-2025 | 93.3% | glm-5.1 | not stated | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GLM-5.1 | FrontierMath Tiers 1-3 (v2) | 24.9% | glm-5.1_none | none | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GLM-5.1 | GPQA Diamond | 89.9% | glm-5.1 | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GLM-5.1 | SimpleQA Verified | 34.0% | glm-5.1 | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GLM-5.1 | SWE-bench Verified (Epoch runs) | 74.2% | glm-5.1 | not stated | share of issues resolved | CC-BY-4.0 | epoch |
|---|
| GLM-5.2 | Arena text: business and management | 1,451 | GLM-5.2 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.2 | Arena text: coding | 1,487 | GLM-5.2 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.2 | Arena text: creative writing | 1,461 | GLM-5.2 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.2 | Arena text: following instructions | 1,466 | GLM-5.2 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.2 | Arena text: maths | 1,484 | GLM-5.2 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.2 | Arena text: overall preference | 1,470 | GLM-5.2 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.2 | Arena WebDev: building web apps | 1,606 | GLM-5.2 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.2 | OTIS Mock AIME 2024-2025 | 86.4% | glm-5.2_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GLM-5.2 | FrontierCode (reported by Cognition) | 24.5% | glm-5.2_none | none, mini-swe-agent | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| GLM-5.2 | FrontierMath Tiers 1-3 (v2) | 59.2% | glm-5.2_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GLM-5.2 | GPQA Diamond | 91.9% | glm-5.2_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GLM-5.2 | SimpleQA Verified | 34.2% | glm-5.2_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GLM-5.2 | SWE-bench Verified (Epoch runs) | 78.7% | glm-5.2_max | max | share of issues resolved | CC-BY-4.0 | epoch |
|---|
| GLM-5.3 | Arena text: business and management | 1,461 | GLM-5.3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.3 | Arena text: coding | 1,497 | GLM-5.3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.3 | Arena text: creative writing | 1,459 | GLM-5.3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.3 | Arena text: following instructions | 1,481 | GLM-5.3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.3 | Arena text: maths | 1,497 | GLM-5.3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.3 | Arena text: overall preference | 1,474 | GLM-5.3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.3 | Arena WebDev: building web apps | 1,623 | GLM-5.3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.3 | OTIS Mock AIME 2024-2025 | 91.1% | glm-5.3_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GLM-5.3 | FrontierCode (reported by Cognition) | 40.1% | glm-5.3_max | max, devin-cli | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| GLM-5.3 | FrontierMath Tiers 1-3 (v2) | 68.8% | glm-5.3_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GLM-5.3 | GPQA Diamond | 90.9% | glm-5.3_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GLM-5.3 | SimpleQA Verified | 41.0% | glm-5.3_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GLM-5.3-Flash | Arena text: business and management | 1,461 | GLM-5.3-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.3-Flash | Arena text: coding | 1,509 | GLM-5.3-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.3-Flash | Arena text: creative writing | 1,441 | GLM-5.3-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.3-Flash | Arena text: following instructions | 1,481 | GLM-5.3-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.3-Flash | Arena text: maths | 1,502 | GLM-5.3-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.3-Flash | Arena text: overall preference | 1,473 | GLM-5.3-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.3-Flash | Arena WebDev: building web apps | 1,607 | GLM-5.3-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GLM-5.3-Flash | OTIS Mock AIME 2024-2025 | 93.9% | glm-5.3-flash_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GLM-5.3-Flash | FrontierCode (reported by Cognition) | 31.8% | glm-5.3-flash_max | max, devin-cli | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| GLM-5.3-Flash | FrontierMath Tiers 1-3 (v2) | 55.8% | glm-5.3-flash_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GLM-5.3-Flash | GPQA Diamond | 90.2% | glm-5.3-flash_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-5.1 | Arena text: business and management | 1,431 | GPT-5.1 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.1 | Arena text: coding | 1,453 | GPT-5.1 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.1 | Arena text: creative writing | 1,427 | GPT-5.1 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.1 | Arena text: following instructions | 1,443 | GPT-5.1 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.1 | Arena text: maths | 1,446 | GPT-5.1 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.1 | Arena text: overall preference | 1,443 | GPT-5.1 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.1 | Arena WebDev: building web apps | 1,396 | GPT-5.1 (Medium) | Medium | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.4 | Arena text: business and management | 1,475 | GPT-5.4 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.4 | Arena text: coding | 1,496 | GPT-5.4 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.4 | Arena text: creative writing | 1,439 | GPT-5.4 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.4 | Arena text: following instructions | 1,469 | GPT-5.4 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.4 | Arena text: maths | 1,488 | GPT-5.4 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.4 | Arena text: overall preference | 1,469 | GPT-5.4 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.4 | Arena WebDev: building web apps | 1,401 | GPT-5.4 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.5 | Arena text: business and management | 1,478 | GPT-5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.5 | Arena text: coding | 1,494 | GPT-5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.5 | Arena text: creative writing | 1,455 | GPT-5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.5 | Arena text: following instructions | 1,479 | GPT-5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.5 | Arena text: maths | 1,478 | GPT-5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.5 | Arena text: overall preference | 1,472 | GPT-5.5 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.5 | OTIS Mock AIME 2024-2025 | 84.4% | gpt-5.5_low | low | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-5.5 | FrontierCode (reported by Cognition) | 43.0% | gpt-5.5_unknown | xhigh, codex | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| GPT-5.5 | FrontierMath Tiers 1-3 (v2) | 85.3% | gpt-5.5_xhigh | xhigh | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-5.5 | GPQA Diamond | 90.7% | gpt-5.5_low | low | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-5.5 | SimpleQA Verified | 63.0% | gpt-5.5_xhigh | xhigh | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-5.5 Instant | Arena text: business and management | 1,410 | GPT-5.5 Instant | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.5 Instant | Arena text: coding | 1,433 | GPT-5.5 Instant | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.5 Instant | Arena text: creative writing | 1,418 | GPT-5.5 Instant | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.5 Instant | Arena text: following instructions | 1,406 | GPT-5.5 Instant | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.5 Instant | Arena text: maths | 1,420 | GPT-5.5 Instant | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.5 Instant | Arena text: overall preference | 1,419 | GPT-5.5 Instant | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.5 Instant | OTIS Mock AIME 2024-2025 | 68.1% | gpt-5.5-instant | not stated | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-5.5 Instant | FrontierMath Tiers 1-3 (v2) | 26.3% | gpt-5.5-instant | not stated | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-5.5 Instant | GPQA Diamond | 82.5% | gpt-5.5-instant | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Luna | Arena text: business and management | 1,439 | GPT-5.6 Luna (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Luna | Arena text: coding | 1,466 | GPT-5.6 Luna (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Luna | Arena text: creative writing | 1,396 | GPT-5.6 Luna (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Luna | Arena text: following instructions | 1,438 | GPT-5.6 Luna (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Luna | Arena text: maths | 1,459 | GPT-5.6 Luna (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Luna | Arena text: overall preference | 1,431 | GPT-5.6 Luna (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Luna | OTIS Mock AIME 2024-2025 | 98.3% | gpt-5.6-luna_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Luna | FrontierCode (reported by Cognition) | 39.8% | gpt-5.6-luna_unknown | max, codex | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Luna | FrontierMath Tiers 1-3 (v2) | 82.1% | gpt-5.6-luna_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Luna | GPQA Diamond | 91.6% | gpt-5.6-luna_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Luna | SimpleQA Verified | 41.0% | gpt-5.6-luna_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Sol | Arena text: business and management | 1,456 | GPT-5.6 Sol (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Sol | Arena text: coding | 1,495 | GPT-5.6 Sol (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Sol | Arena text: creative writing | 1,450 | GPT-5.6 Sol (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Sol | Arena text: following instructions | 1,483 | GPT-5.6 Sol (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Sol | Arena text: maths | 1,472 | GPT-5.6 Sol (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Sol | Arena text: overall preference | 1,457 | GPT-5.6 Sol (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Sol | OTIS Mock AIME 2024-2025 | 100.0% | gpt-5.6-sol_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Sol | FrontierCode (reported by Cognition) | 47.5% | gpt-5.6-sol_unknown | max, codex | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Sol | FrontierMath Tiers 1-3 (v2) | 89.1% | gpt-5.6-sol_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Sol | GPQA Diamond | 93.5% | gpt-5.6-sol_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Sol | SimpleQA Verified | 69.7% | gpt-5.6-sol_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Terra | Arena text: business and management | 1,447 | GPT-5.6 Terra (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Terra | Arena text: coding | 1,485 | GPT-5.6 Terra (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Terra | Arena text: creative writing | 1,413 | GPT-5.6 Terra (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Terra | Arena text: following instructions | 1,455 | GPT-5.6 Terra (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Terra | Arena text: maths | 1,464 | GPT-5.6 Terra (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Terra | Arena text: overall preference | 1,448 | GPT-5.6 Terra (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-5.6 Terra | OTIS Mock AIME 2024-2025 | 99.7% | gpt-5.6-terra_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Terra | FrontierCode (reported by Cognition) | 41.3% | gpt-5.6-terra_unknown | max, codex | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Terra | FrontierMath Tiers 1-3 (v2) | 86.0% | gpt-5.6-terra_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Terra | GPQA Diamond | 93.3% | gpt-5.6-terra_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-5.6 Terra | SimpleQA Verified | 43.2% | gpt-5.6-terra_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-6.1 Sol | Arena text: business and management | 1,449 | GPT-6.1 Sol (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6.1 Sol | Arena text: coding | 1,486 | GPT-6.1 Sol (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6.1 Sol | Arena text: creative writing | 1,425 | GPT-6.1 Sol (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6.1 Sol | Arena text: following instructions | 1,467 | GPT-6.1 Sol (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6.1 Sol | Arena text: maths | 1,486 | GPT-6.1 Sol (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6.1 Sol | Arena text: overall preference | 1,446 | GPT-6.1 Sol (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6.1 Sol | Arena WebDev: building web apps | 1,751 | GPT-6.1 Sol (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6.1 Sol | OTIS Mock AIME 2024-2025 | 100.0% | gpt-6.1-sol_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-6.1 Sol | FrontierCode (reported by Cognition) | 50.2% | gpt-6.1-sol_medium | medium, codex | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| GPT-6.1 Sol | FrontierMath Tiers 1-3 (v2) | 93.7% | gpt-6.1-sol_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-6.1 Sol | GPQA Diamond | 95.4% | gpt-6.1-sol_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-6.1 Sol | SimpleQA Verified | 73.9% | gpt-6.1-sol_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-6 Astra | Arena text: business and management | 1,438 | GPT-6 Astra (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Astra | Arena text: coding | 1,488 | GPT-6 Astra (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Astra | Arena text: creative writing | 1,420 | GPT-6 Astra (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Astra | Arena text: following instructions | 1,454 | GPT-6 Astra (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Astra | Arena text: maths | 1,468 | GPT-6 Astra (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Astra | Arena text: overall preference | 1,441 | GPT-6 Astra (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Astra | Arena WebDev: building web apps | 1,773 | GPT-6 Astra (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Astra | OTIS Mock AIME 2024-2025 | 100.0% | gpt-6-astra_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-6 Astra | FrontierCode (reported by Cognition) | 53.3% | gpt-6-astra_max | max, codex | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| GPT-6 Astra | FrontierMath Tiers 1-3 (v2) | 93.7% | gpt-6-astra_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-6 Astra | GPQA Diamond | 95.8% | gpt-6-astra_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-6 Astra | SimpleQA Verified | 75.6% | gpt-6-astra_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-6 Luna | Arena text: business and management | 1,384 | GPT-6 Luna (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Luna | Arena text: coding | 1,437 | GPT-6 Luna (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Luna | Arena text: creative writing | 1,362 | GPT-6 Luna (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Luna | Arena text: following instructions | 1,406 | GPT-6 Luna (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Luna | Arena text: maths | 1,410 | GPT-6 Luna (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Luna | Arena text: overall preference | 1,389 | GPT-6 Luna (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Luna | Arena WebDev: building web apps | 1,579 | GPT-6 Luna (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Luna | OTIS Mock AIME 2024-2025 | 98.9% | gpt-6-luna_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-6 Luna | FrontierCode (reported by Cognition) | 42.4% | gpt-6-luna_max | max, codex | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| GPT-6 Luna | FrontierMath Tiers 1-3 (v2) | 78.9% | gpt-6-luna_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-6 Luna | GPQA Diamond | 90.5% | gpt-6-luna_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-6 Luna | SimpleQA Verified | 41.4% | gpt-6-luna_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-6 Sol | Arena text: business and management | 1,381 | GPT-6 Sol (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Sol | Arena text: coding | 1,447 | GPT-6 Sol (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Sol | Arena text: creative writing | 1,378 | GPT-6 Sol (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Sol | Arena text: following instructions | 1,411 | GPT-6 Sol (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Sol | Arena text: maths | 1,404 | GPT-6 Sol (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Sol | Arena text: overall preference | 1,397 | GPT-6 Sol (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Sol | Arena WebDev: building web apps | 1,678 | GPT-6 Sol (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| GPT-6 Sol | OTIS Mock AIME 2024-2025 | 100.0% | gpt-6-sol_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-6 Sol | FrontierCode (reported by Cognition) | 49.3% | gpt-6-sol_max | max, codex | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| GPT-6 Sol | FrontierMath Tiers 1-3 (v2) | 89.8% | gpt-6-sol_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| GPT-6 Sol | GPQA Diamond | 94.3% | gpt-6-sol_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| GPT-6 Sol | SimpleQA Verified | 60.7% | gpt-6-sol_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Grok 4.20 Beta 1 | Arena text: business and management | 1,425 | Grok 4.20 Beta 1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Beta 1 | Arena text: coding | 1,447 | Grok 4.20 Beta 1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Beta 1 | Arena text: creative writing | 1,438 | Grok 4.20 Beta 1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Beta 1 | Arena text: following instructions | 1,415 | Grok 4.20 Beta 1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Beta 1 | Arena text: maths | 1,432 | Grok 4.20 Beta 1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Beta 1 | Arena text: overall preference | 1,444 | Grok 4.20 Beta 1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Beta (Reasoning, 0309) | Arena text: business and management | 1,432 | Grok 4.20 Beta (Reasoning, 0309) | Reasoning, 0309 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Beta (Reasoning, 0309) | Arena text: coding | 1,459 | Grok 4.20 Beta (Reasoning, 0309) | Reasoning, 0309 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Beta (Reasoning, 0309) | Arena text: creative writing | 1,432 | Grok 4.20 Beta (Reasoning, 0309) | Reasoning, 0309 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Beta (Reasoning, 0309) | Arena text: following instructions | 1,421 | Grok 4.20 Beta (Reasoning, 0309) | Reasoning, 0309 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Beta (Reasoning, 0309) | Arena text: maths | 1,455 | Grok 4.20 Beta (Reasoning, 0309) | Reasoning, 0309 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Beta (Reasoning, 0309) | Arena text: overall preference | 1,451 | Grok 4.20 Beta (Reasoning, 0309) | Reasoning, 0309 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Beta (Reasoning, 0309) | Arena WebDev: building web apps | 1,375 | Grok 4.20 Beta (Reasoning, 0309) | Reasoning, 0309 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Multi-Agent Beta (0309) | Arena text: business and management | 1,423 | Grok 4.20 Multi-Agent Beta (0309) | 0309 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Multi-Agent Beta (0309) | Arena text: coding | 1,457 | Grok 4.20 Multi-Agent Beta (0309) | 0309 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Multi-Agent Beta (0309) | Arena text: creative writing | 1,436 | Grok 4.20 Multi-Agent Beta (0309) | 0309 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Multi-Agent Beta (0309) | Arena text: following instructions | 1,420 | Grok 4.20 Multi-Agent Beta (0309) | 0309 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Multi-Agent Beta (0309) | Arena text: maths | 1,441 | Grok 4.20 Multi-Agent Beta (0309) | 0309 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.20 Multi-Agent Beta (0309) | Arena text: overall preference | 1,451 | Grok 4.20 Multi-Agent Beta (0309) | 0309 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.3 | Arena text: business and management | 1,381 | Grok 4.3 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.3 | Arena text: coding | 1,414 | Grok 4.3 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.3 | Arena text: creative writing | 1,380 | Grok 4.3 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.3 | Arena text: following instructions | 1,366 | Grok 4.3 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.3 | Arena text: maths | 1,387 | Grok 4.3 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.3 | Arena text: overall preference | 1,397 | Grok 4.3 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.3 | Arena WebDev: building web apps | 1,358 | Grok 4.3 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.5 | Arena text: business and management | 1,441 | Grok 4.5 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.5 | Arena text: coding | 1,474 | Grok 4.5 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.5 | Arena text: creative writing | 1,442 | Grok 4.5 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.5 | Arena text: following instructions | 1,445 | Grok 4.5 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.5 | Arena text: maths | 1,459 | Grok 4.5 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.5 | Arena text: overall preference | 1,448 | Grok 4.5 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.5 | Arena WebDev: building web apps | 1,553 | Grok 4.5 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.5 | OTIS Mock AIME 2024-2025 | 97.8% | grok-4.5_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Grok 4.5 | FrontierCode (reported by Cognition) | 42.4% | grok-4.5_unknown | high, grok-build | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Grok 4.5 | FrontierMath Tiers 1-3 (v2) | 57.2% | grok-4.5_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Grok 4.5 | GPQA Diamond | 93.4% | grok-4.5_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Grok 4.5 | SimpleQA Verified | 48.3% | grok-4.5_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Grok 4.6 | Arena text: business and management | 1,416 | Grok 4.6 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.6 | Arena text: coding | 1,464 | Grok 4.6 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.6 | Arena text: creative writing | 1,426 | Grok 4.6 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.6 | Arena text: following instructions | 1,431 | Grok 4.6 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.6 | Arena text: maths | 1,415 | Grok 4.6 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.6 | Arena text: overall preference | 1,429 | Grok 4.6 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.6 | Arena WebDev: building web apps | 1,616 | Grok 4.6 (High) | High | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.6 | OTIS Mock AIME 2024-2025 | 99.2% | grok-4.6_xhigh | xhigh | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Grok 4.6 | FrontierCode (reported by Cognition) | 48.0% | grok-4.6_unknown | high, grok-build | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Grok 4.6 | FrontierMath Tiers 1-3 (v2) | 66.0% | grok-4.6_xhigh | xhigh | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Grok 4.6 | GPQA Diamond | 93.2% | grok-4.6_xhigh | xhigh | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Grok 4.6 | SimpleQA Verified | 48.9% | grok-4.6_xhigh | xhigh | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Grok 4.7 | Arena text: business and management | 1,390 | Grok 4.7 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.7 | Arena text: coding | 1,428 | Grok 4.7 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.7 | Arena text: creative writing | 1,381 | Grok 4.7 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.7 | Arena text: following instructions | 1,406 | Grok 4.7 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.7 | Arena text: maths | 1,413 | Grok 4.7 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.7 | Arena text: overall preference | 1,398 | Grok 4.7 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.7 | Arena WebDev: building web apps | 1,637 | Grok 4.7 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Grok 4.7 | OTIS Mock AIME 2024-2025 | 98.1% | grok-4.7_xhigh | xhigh | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Grok 4.7 | FrontierCode (reported by Cognition) | 47.6% | grok-4.7_unknown | high, grok-build | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Grok 4.7 | FrontierMath Tiers 1-3 (v2) | 53.0% | grok-4.7_xhigh | xhigh | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Grok 4.7 | GPQA Diamond | 92.7% | grok-4.7_xhigh | xhigh | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Grok 4.7 | SimpleQA Verified | 56.0% | grok-4.7_xhigh | xhigh | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Inkling-Small | Arena text: business and management | 1,414 | Inkling-Small | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Inkling-Small | Arena text: coding | 1,451 | Inkling-Small | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Inkling-Small | Arena text: creative writing | 1,336 | Inkling-Small | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Inkling-Small | Arena text: following instructions | 1,401 | Inkling-Small | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Inkling-Small | Arena text: maths | 1,458 | Inkling-Small | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Inkling-Small | Arena text: overall preference | 1,414 | Inkling-Small | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Inkling-Small | Arena WebDev: building web apps | 1,409 | Inkling-Small | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Inkling-Small | OTIS Mock AIME 2024-2025 | 90.0% | inkling-small_xhigh | xhigh | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Inkling-Small | FrontierMath Tiers 1-3 (v2) | 46.3% | inkling-small_xhigh | xhigh | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Inkling-Small | GPQA Diamond | 88.5% | inkling-small_xhigh | xhigh | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Inkling-Small | SimpleQA Verified | 19.1% | inkling-small_xhigh | xhigh | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Kimi K2.5 | Arena text: business and management | 1,427 | Kimi K2.5 (Thinking) | Thinking | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K2.5 | Arena text: coding | 1,474 | Kimi K2.5 (Thinking) | Thinking | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K2.5 | Arena text: creative writing | 1,422 | Kimi K2.5 (Thinking) | Thinking | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K2.5 | Arena text: following instructions | 1,431 | Kimi K2.5 (Thinking) | Thinking | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K2.5 | Arena text: maths | 1,470 | Kimi K2.5 (Thinking) | Thinking | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K2.5 | Arena text: overall preference | 1,445 | Kimi K2.5 (Thinking) | Thinking | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K2.5 | Arena WebDev: building web apps | 1,436 | Kimi K2.5 (Thinking) | Thinking | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K2.5 | SimpleQA Verified | 34.3% | kimi-k2.5 | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Kimi K2.5 | SWE-bench Verified (Epoch runs) | 73.8% | kimi-k2.5 | not stated | share of issues resolved | CC-BY-4.0 | epoch |
|---|
| Kimi K2.6 | Arena text: business and management | 1,445 | Kimi K2.6 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K2.6 | Arena text: coding | 1,489 | Kimi K2.6 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K2.6 | Arena text: creative writing | 1,433 | Kimi K2.6 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K2.6 | Arena text: following instructions | 1,451 | Kimi K2.6 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K2.6 | Arena text: maths | 1,474 | Kimi K2.6 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K2.6 | Arena text: overall preference | 1,456 | Kimi K2.6 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K2.6 | Arena WebDev: building web apps | 1,509 | Kimi K2.6 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K2.6 | OTIS Mock AIME 2024-2025 | 96.1% | kimi-k2.6 | not stated | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Kimi K2.6 | FrontierMath Tiers 1-3 (v2) | 57.2% | kimi-k2.6 | not stated | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Kimi K2.6 | GPQA Diamond | 90.8% | kimi-k2.6 | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Kimi K2.6 | SimpleQA Verified | 34.9% | kimi-k2.6 | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Kimi K2.6 | SWE-bench Verified (Epoch runs) | 76.7% | kimi-k2.6 | not stated | share of issues resolved | CC-BY-4.0 | epoch |
|---|
| Kimi K3 | Arena text: business and management | 1,469 | Kimi K3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K3 | Arena text: coding | 1,508 | Kimi K3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K3 | Arena text: creative writing | 1,457 | Kimi K3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K3 | Arena text: following instructions | 1,485 | Kimi K3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K3 | Arena text: maths | 1,492 | Kimi K3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K3 | Arena text: overall preference | 1,476 | Kimi K3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K3 | Arena WebDev: building web apps | 1,656 | Kimi K3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Kimi K3 | OTIS Mock AIME 2024-2025 | 97.2% | kimi-k3_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Kimi K3 | FrontierCode (reported by Cognition) | 44.2% | kimi-k3_unknown | none, mini-swe-agent | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Kimi K3 | FrontierMath Tiers 1-3 (v2) | 72.2% | kimi-k3_max | max | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Kimi K3 | GPQA Diamond | 93.1% | kimi-k3_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Kimi K3 | SimpleQA Verified | 50.6% | kimi-k3_max | max | share of questions correct | CC-BY-4.0 | epoch |
|---|
| MiMo-V2.5-Pro | Arena text: business and management | 1,467 | MiMo-V2.5-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.5-Pro | Arena text: coding | 1,501 | MiMo-V2.5-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.5-Pro | Arena text: creative writing | 1,438 | MiMo-V2.5-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.5-Pro | Arena text: following instructions | 1,475 | MiMo-V2.5-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.5-Pro | Arena text: maths | 1,481 | MiMo-V2.5-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.5-Pro | Arena text: overall preference | 1,464 | MiMo-V2.5-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.5-Pro | Arena WebDev: building web apps | 1,480 | MiMo-V2.5-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.6-Flash | Arena text: business and management | 1,450 | MiMo-V2.6-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.6-Flash | Arena text: coding | 1,498 | MiMo-V2.6-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.6-Flash | Arena text: creative writing | 1,405 | MiMo-V2.6-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.6-Flash | Arena text: following instructions | 1,449 | MiMo-V2.6-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.6-Flash | Arena text: maths | 1,468 | MiMo-V2.6-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.6-Flash | Arena text: overall preference | 1,451 | MiMo-V2.6-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.6-Flash | Arena WebDev: building web apps | 1,637 | MiMo-V2.6-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.6-Pro | Arena text: business and management | 1,468 | MiMo-V2.6-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.6-Pro | Arena text: coding | 1,534 | MiMo-V2.6-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.6-Pro | Arena text: creative writing | 1,464 | MiMo-V2.6-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.6-Pro | Arena text: following instructions | 1,492 | MiMo-V2.6-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.6-Pro | Arena text: maths | 1,501 | MiMo-V2.6-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.6-Pro | Arena text: overall preference | 1,491 | MiMo-V2.6-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| MiMo-V2.6-Pro | Arena WebDev: building web apps | 1,625 | MiMo-V2.6-Pro | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Mistral Large 4 | Arena text: business and management | 1,439 | Mistral Large 4 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Mistral Large 4 | Arena text: coding | 1,478 | Mistral Large 4 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Mistral Large 4 | Arena text: creative writing | 1,361 | Mistral Large 4 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Mistral Large 4 | Arena text: following instructions | 1,419 | Mistral Large 4 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Mistral Large 4 | Arena text: maths | 1,468 | Mistral Large 4 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Mistral Large 4 | Arena text: overall preference | 1,430 | Mistral Large 4 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Mistral Large 4 | Arena WebDev: building web apps | 1,544 | Mistral Large 4 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Mistral Large 4 | OTIS Mock AIME 2024-2025 | 98.3% | mistral-large-4-0_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Mistral Large 4 | FrontierCode (reported by Cognition) | 39.4% | mistral-large-4-0_high | high, chisel | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Mistral Large 4 | FrontierMath Tiers 1-3 (v2) | 58.6% | mistral-large-4-0_high | high | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Mistral Large 4 | GPQA Diamond | 88.1% | mistral-large-4-0_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Mistral Large 4 | SimpleQA Verified | 20.0% | mistral-large-4-0_high | high | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Muse Spark | Arena text: business and management | 1,462 | Muse Spark | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark | Arena text: coding | 1,481 | Muse Spark | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark | Arena text: creative writing | 1,459 | Muse Spark | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark | Arena text: following instructions | 1,442 | Muse Spark | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark | Arena text: maths | 1,456 | Muse Spark | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark | Arena text: overall preference | 1,474 | Muse Spark | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark | OTIS Mock AIME 2024-2025 | 88.9% | muse-spark | not stated | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Muse Spark | GPQA Diamond | 89.8% | muse-spark | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Muse Spark 1.1 | Arena text: business and management | 1,470 | Muse Spark 1.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.1 | Arena text: coding | 1,498 | Muse Spark 1.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.1 | Arena text: creative writing | 1,435 | Muse Spark 1.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.1 | Arena text: following instructions | 1,457 | Muse Spark 1.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.1 | Arena text: maths | 1,482 | Muse Spark 1.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.1 | Arena text: overall preference | 1,478 | Muse Spark 1.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.1 | Arena WebDev: building web apps | 1,542 | Muse Spark 1.1 | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.1 | SimpleQA Verified | 57.8% | muse-spark-1.1 | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Muse Spark 1.2 | Arena text: business and management | 1,484 | Muse Spark 1.2 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.2 | Arena text: coding | 1,506 | Muse Spark 1.2 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.2 | Arena text: creative writing | 1,437 | Muse Spark 1.2 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.2 | Arena text: following instructions | 1,465 | Muse Spark 1.2 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.2 | Arena text: maths | 1,485 | Muse Spark 1.2 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.2 | Arena text: overall preference | 1,483 | Muse Spark 1.2 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.2 | Arena WebDev: building web apps | 1,534 | Muse Spark 1.2 (xHigh) | xHigh | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.2 | SimpleQA Verified | 60.3% | muse-spark-1.2_xhigh | xhigh | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Muse Spark 1.3 | Arena text: business and management | 1,486 | Muse Spark 1.3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.3 | Arena text: coding | 1,509 | Muse Spark 1.3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.3 | Arena text: creative writing | 1,464 | Muse Spark 1.3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.3 | Arena text: following instructions | 1,478 | Muse Spark 1.3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.3 | Arena text: maths | 1,501 | Muse Spark 1.3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.3 | Arena text: overall preference | 1,491 | Muse Spark 1.3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.3 | Arena WebDev: building web apps | 1,656 | Muse Spark 1.3 (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Spark 1.3 | OTIS Mock AIME 2024-2025 | 99.2% | muse-spark-1.3_xhigh | xhigh | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Muse Spark 1.3 | FrontierMath Tiers 1-3 (v2) | 74.4% | muse-spark-1.3_xhigh | xhigh | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Muse Wonder | Arena text: business and management | 1,487 | Muse Wonder (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Wonder | Arena text: coding | 1,518 | Muse Wonder (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Wonder | Arena text: creative writing | 1,488 | Muse Wonder (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Wonder | Arena text: following instructions | 1,494 | Muse Wonder (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Wonder | Arena text: maths | 1,477 | Muse Wonder (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Wonder | Arena text: overall preference | 1,499 | Muse Wonder (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| Muse Wonder | Arena WebDev: building web apps | 1,652 | Muse Wonder (Max) | Max | Arena score | CC-BY-4.0 | lmarena |
|---|
| NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4) | Arena text: business and management | 1,435 | NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4) | NVFP4 | Arena score | CC-BY-4.0 | lmarena |
|---|
| NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4) | Arena text: coding | 1,468 | NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4) | NVFP4 | Arena score | CC-BY-4.0 | lmarena |
|---|
| NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4) | Arena text: creative writing | 1,398 | NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4) | NVFP4 | Arena score | CC-BY-4.0 | lmarena |
|---|
| NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4) | Arena text: following instructions | 1,418 | NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4) | NVFP4 | Arena score | CC-BY-4.0 | lmarena |
|---|
| NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4) | Arena text: maths | 1,458 | NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4) | NVFP4 | Arena score | CC-BY-4.0 | lmarena |
|---|
| NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4) | Arena text: overall preference | 1,445 | NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4) | NVFP4 | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-397B-A17B | Arena text: business and management | 1,435 | Qwen3.5-397B-A17B | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-397B-A17B | Arena text: coding | 1,464 | Qwen3.5-397B-A17B | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-397B-A17B | Arena text: creative writing | 1,402 | Qwen3.5-397B-A17B | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-397B-A17B | Arena text: following instructions | 1,423 | Qwen3.5-397B-A17B | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-397B-A17B | Arena text: maths | 1,454 | Qwen3.5-397B-A17B | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-397B-A17B | Arena text: overall preference | 1,438 | Qwen3.5-397B-A17B | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-397B-A17B | Arena WebDev: building web apps | 1,400 | Qwen3.5-397B-A17B | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-397B-A17B | OTIS Mock AIME 2024-2025 | 82.2% | qwen3.5-397b-a17b_none | none | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Qwen3.5-397B-A17B | FrontierMath Tiers 1-3 (v2) | 31.2% | qwen3.5-397b-a17b_none | none | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Qwen3.5-397B-A17B | GPQA Diamond | 86.4% | qwen3.5-397b-a17b_none | none | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Qwen3.5-Flash | Arena text: business and management | 1,394 | Qwen3.5-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-Flash | Arena text: coding | 1,412 | Qwen3.5-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-Flash | Arena text: creative writing | 1,343 | Qwen3.5-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-Flash | Arena text: following instructions | 1,375 | Qwen3.5-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-Flash | Arena text: maths | 1,408 | Qwen3.5-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-Flash | Arena text: overall preference | 1,397 | Qwen3.5-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-Flash | Arena WebDev: building web apps | 1,244 | Qwen3.5-Flash | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-Max-Preview | Arena text: business and management | 1,463 | Qwen3.5-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-Max-Preview | Arena text: coding | 1,487 | Qwen3.5-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-Max-Preview | Arena text: creative writing | 1,464 | Qwen3.5-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-Max-Preview | Arena text: following instructions | 1,467 | Qwen3.5-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-Max-Preview | Arena text: maths | 1,474 | Qwen3.5-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.5-Max-Preview | Arena text: overall preference | 1,470 | Qwen3.5-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Max-Preview | Arena text: business and management | 1,427 | Qwen3.6-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Max-Preview | Arena text: coding | 1,471 | Qwen3.6-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Max-Preview | Arena text: creative writing | 1,435 | Qwen3.6-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Max-Preview | Arena text: following instructions | 1,439 | Qwen3.6-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Max-Preview | Arena text: maths | 1,464 | Qwen3.6-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Max-Preview | Arena text: overall preference | 1,447 | Qwen3.6-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Max-Preview | Arena WebDev: building web apps | 1,482 | Qwen3.6-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Max-Preview | OTIS Mock AIME 2024-2025 | 91.1% | qwen3.6-max-preview | not stated | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Qwen3.6-Max-Preview | GPQA Diamond | 87.4% | qwen3.6-max-preview | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Qwen3.6-Max-Preview | SimpleQA Verified | 52.0% | qwen3.6-max-preview | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Qwen3.6-Max-Preview | SWE-bench Verified (Epoch runs) | 76.7% | qwen3.6-max-preview | not stated | share of issues resolved | CC-BY-4.0 | epoch |
|---|
| Qwen3.6-Plus | Arena text: business and management | 1,430 | Qwen3.6-Plus | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Plus | Arena text: coding | 1,467 | Qwen3.6-Plus | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Plus | Arena text: creative writing | 1,403 | Qwen3.6-Plus | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Plus | Arena text: following instructions | 1,424 | Qwen3.6-Plus | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Plus | Arena text: maths | 1,449 | Qwen3.6-Plus | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Plus | Arena text: overall preference | 1,436 | Qwen3.6-Plus | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Plus | Arena WebDev: building web apps | 1,462 | Qwen3.6-Plus | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.6-Plus | OTIS Mock AIME 2024-2025 | 93.3% | qwen3.6-plus | not stated | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Qwen3.6-Plus | FrontierMath Tiers 1-3 (v2) | 32.3% | qwen3.6-plus_none | none | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Qwen3.6-Plus | GPQA Diamond | 88.4% | qwen3.6-plus | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Qwen3.6-Plus | SimpleQA Verified | 44.1% | qwen3.6-plus | not stated | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Qwen3.6-Plus | SWE-bench Verified (Epoch runs) | 57.9% | qwen3.6-plus | not stated | share of issues resolved | CC-BY-4.0 | epoch |
|---|
| Qwen3.7-Max-Preview | Arena text: business and management | 1,446 | Qwen3.7-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.7-Max-Preview | Arena text: coding | 1,497 | Qwen3.7-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.7-Max-Preview | Arena text: creative writing | 1,446 | Qwen3.7-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.7-Max-Preview | Arena text: following instructions | 1,460 | Qwen3.7-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.7-Max-Preview | Arena text: maths | 1,486 | Qwen3.7-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.7-Max-Preview | Arena text: overall preference | 1,475 | Qwen3.7-Max-Preview | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.7-Plus | Arena text: business and management | 1,435 | Qwen3.7-Plus | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.7-Plus | Arena text: coding | 1,473 | Qwen3.7-Plus | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.7-Plus | Arena text: creative writing | 1,438 | Qwen3.7-Plus | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.7-Plus | Arena text: following instructions | 1,440 | Qwen3.7-Plus | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.7-Plus | Arena text: maths | 1,467 | Qwen3.7-Plus | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.7-Plus | Arena text: overall preference | 1,455 | Qwen3.7-Plus | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.7-Plus | OTIS Mock AIME 2024-2025 | 80.0% | qwen3.7-plus_none | none | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Qwen3.7-Plus | FrontierCode (reported by Cognition) | 10.2% | qwen3.7-plus | none, mini-swe-agent | mean score over 5 attempts | CC-BY-4.0 | epoch |
|---|
| Qwen3.7-Plus | FrontierMath Tiers 1-3 (v2) | 34.4% | qwen3.7-plus_none | none | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Qwen3.7-Plus | GPQA Diamond | 81.8% | qwen3.7-plus_none | none | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Qwen3.8-Max | Arena text: business and management | 1,480 | Qwen3.8-Max | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.8-Max | Arena text: coding | 1,504 | Qwen3.8-Max | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.8-Max | Arena text: creative writing | 1,478 | Qwen3.8-Max | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.8-Max | Arena text: following instructions | 1,480 | Qwen3.8-Max | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.8-Max | Arena text: maths | 1,506 | Qwen3.8-Max | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.8-Max | Arena text: overall preference | 1,483 | Qwen3.8-Max | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.8-Max | Arena WebDev: building web apps | 1,672 | Qwen3.8-Max | not stated | Arena score | CC-BY-4.0 | lmarena |
|---|
| Qwen3.8-Max | OTIS Mock AIME 2024-2025 | 99.4% | qwen3.8-max_xhigh | xhigh | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Qwen3.8-Max | FrontierMath Tiers 1-3 (v2) | 74.7% | qwen3.8-max_xhigh | xhigh | share of problems solved | CC-BY-4.0 | epoch |
|---|
| Qwen3.8-Max | GPQA Diamond | 92.7% | qwen3.8-max_xhigh | xhigh | share of questions correct | CC-BY-4.0 | epoch |
|---|
| Qwen3.8-Max | SimpleQA Verified | 45.8% | qwen3.8-max_xhigh | xhigh | share of questions correct | CC-BY-4.0 | epoch |
|---|