Qwen3.8-Max
Made by Alibaba, released 2 Aug 2026. Epoch AI lists its access as “API access”. Appears strongest at coding and writing among the models listed here.
In short
What it does well
- Coding: Top tier (16th of 77 listed models)
- Writing: Top tier (13th of 77 listed models)
- Maths: Strong (17th of 43 listed models)
Where it falls short
- Knowledge: Middle (28th of 43 listed models)
What we don't know yet
- Fixing real software bugs: Not enough data
- Difficult coding tasks: Not enough data
The four ratings
- Top tier
- Strong
- Middle
- Behind the leaders
- Not enough data
Ratings compare only the 77 models listed here, in quarters, best first. “Not enough data” means no result yet. It is never a low score.
- Writing
- Top tierHow often people preferred its answers to writing prompts. 1,478 on Arena's scale · Source for Qwen3.8-Max writing score (lmarena)
- Coding
- Top tierHow often people preferred its answers to coding questions. 1,504 on Arena's scale · Source for Qwen3.8-Max coding score (lmarena)
- Maths
- StrongHow many very hard maths problems it solved. 74.7% · Source for Qwen3.8-Max maths score (epoch)
- Knowledge
- MiddleHow many short factual questions it answered correctly. 45.8% · Source for Qwen3.8-Max knowledge score (epoch)
Evidence completeness: Broad evidence, 11 of 13 tests have a result.
Where it is relatively stronger or weaker
This compares each kind of request with the model's own overall position among the 77 listed models, using Arena's votes. It is about performance on different languages, kinds of request and topics, not about political or social bias.
Relatively stronger
No category stands out as clearly stronger than its overall position.
Relatively weaker
No category stands out as clearly weaker than its overall position.
Show every category we can see
| Category | Position | Arena score | Votes | Source |
|---|---|---|---|---|
| Expert-level prompts Kind of request | 17th of 77 | 1,508 95% range 1,497 to 1,518 | 3,496 | Source for Expert-level prompts |
| Hard prompts Kind of request | 13th of 77 | 1,496 95% range 1,491 to 1,502 | 20,978 | Source for Hard prompts |
| Long prompts Kind of request | 14th of 77 | 1,488 95% range 1,482 to 1,495 | 14,028 | Source for Long prompts |
| Longer conversations Kind of request | 11th of 77 | 1,491 95% range 1,482 to 1,500 | 4,825 | Source for Longer conversations |
| Chinese Language | 9th of 76 | 1,540 95% range 1,527 to 1,553 | 2,425 | Source for Chinese |
| English prompts Language | 12th of 77 | 1,493 95% range 1,486 to 1,499 | 12,039 | Source for English prompts |
| French Language | 8th of 72 | 1,503 95% range 1,484 to 1,521 | 1,157 | Source for French |
| German Language | 18th of 55 | 1,475 95% range 1,450 to 1,501 | 500 | Source for German |
| Japanese Language | 18th of 54 | 1,471 95% range 1,445 to 1,498 | 581 | Source for Japanese |
| Korean Language | 6th of 54 | 1,462 95% range 1,438 to 1,486 | 652 | Source for Korean |
| Non-English prompts Language | 18th of 77 | 1,472 95% range 1,467 to 1,478 | 19,871 | Source for Non-English prompts |
| Russian Language | 17th of 77 | 1,484 95% range 1,474 to 1,493 | 4,278 | Source for Russian |
| Spanish Language | 10th of 70 | 1,488 95% range 1,469 to 1,507 | 1,026 | Source for Spanish |
| Entertainment, sport and media Topic | 12th of 77 | 1,465 95% range 1,457 to 1,473 | 7,969 | Source for Entertainment, sport and media |
| Law and government Topic | 14th of 76 | 1,494 95% range 1,481 to 1,506 | 2,352 | Source for Law and government |
| Medicine and health Topic | 7th of 75 | 1,505 95% range 1,491 to 1,518 | 2,113 | Source for Medicine and health |
| Science Topic | 7th of 77 | 1,514 95% range 1,505 to 1,523 | 4,948 | Source for Science |
| Software and IT Topic | 17th of 77 | 1,499 95% range 1,492 to 1,505 | 12,546 | Source for Software and IT |
Bias: what we can and can't see
Our sources do not measure political lean, stereotyping, flattery (agreeing with you too readily) or safety behaviour, so we cannot say how any model does on those. What we can show is above: where a model is relatively stronger or weaker across languages, kinds of request and topics. Also, people voting in Arena can favour answer style, such as length and formatting, and our numbers do not adjust for that. See what AI bias means and what these tests can tell you.
Best-supported use cases
This model is not among the options we recommend for any of the five tasks. That reflects the rules and the models listed, not a verdict on the model.
Where an alternative might be more suitable
- Homework and maths: rated Strong on hard maths problems. Compare it with GPT-6.1 Sol, the best-supported choice for this task.
- Research and checking facts: rated Middle on short factual questions. Compare it with Gemini 4 Argon, the best-supported choice for this task.
What it is
- Maker
- Alibaba
- Release date (Epoch AI)
- 2 Aug 2026
- Access, as listed by Epoch AI
- API access
- Licence, as listed by Arena
- Proprietary
- Maker's country (Epoch AI)
- China
- Last observation in our data
- 28 Oct 2026
- Data last changed
- 29 Oct 2026, 14:36 UTC
A separate 'Qwen3.8-Max (0902)' entry exists in both sources; it is not merged with this one.
We have not verified an official product page for this model, so we do not link to one.
How you can use it
Proprietary. The model's files are not published, so it is generally available only through its maker or partners the maker chooses.
How you can reach it (an app, a website, a developer API or a third-party tool) and what that costs are not covered by our benchmark data. Where we could verify prices from the maker's own pages, they are shown under Price. Epoch AI lists access as “API access”. See models, apps and harnesses and how AI pricing works.
Price
Price not independently verified
We have not been able to check this maker's own pricing page, so we show nothing for it.
We never infer subscription prices from API prices. New to “tokens” and “per million”? See how AI pricing works.
Good to remember
- Benchmarks are samples of tasks. None proves performance in every situation.
- Arena scores come from people choosing between answers. That measures preference, not correctness.
- Ratings are relative to the 77 models listed here. A different set would change them.
- Sources test models at different settings. Each result shows the setting used, and we do not adjust it.
The evidence, for the curious
Show all 13 tests with settings and sources
| Test | Result | Rating | Setting used | Source identifier | Observed | Links |
|---|---|---|---|---|---|---|
| Arena text: overall preference Arena 'text' leaderboard (Bradley-Terry rating, no style control) |
1,483 Arena score 95% range 1,479 to 1,488, 31,915 votes |
Top tier | effort: not stated, bradley-terry rating | Qwen3.8-Max lmarena · text/latest |
28 Oct 2026 | Source for Qwen3.8-Max Arena text: overall preference score (lmarena) |
| Arena text: creative writing Arena 'text' leaderboard, category creative_writing |
1,478 Arena score 95% range 1,470 to 1,487, 6,348 votes |
Top tier | effort: not stated, bradley-terry rating | Qwen3.8-Max lmarena · text/latest |
28 Oct 2026 | Source for Qwen3.8-Max Arena text: creative writing score (lmarena) |
| Arena text: coding Arena 'text' leaderboard, category coding |
1,504 Arena score 95% range 1,496 to 1,511, 8,708 votes |
Top tier | effort: not stated, bradley-terry rating | Qwen3.8-Max lmarena · text/latest |
28 Oct 2026 | Source for Qwen3.8-Max Arena text: coding score (lmarena) |
| Arena text: maths Arena 'text' leaderboard, category math |
1,506 Arena score 95% range 1,491 to 1,521, 1,622 votes |
Top tier | effort: not stated, bradley-terry rating | Qwen3.8-Max lmarena · text/latest |
28 Oct 2026 | Source for Qwen3.8-Max Arena text: maths score (lmarena) |
| Arena text: following instructions Arena 'text' leaderboard, category instruction_following |
1,480 Arena score 95% range 1,474 to 1,487, 11,282 votes |
Top tier | effort: not stated, bradley-terry rating | Qwen3.8-Max lmarena · text/latest |
28 Oct 2026 | Source for Qwen3.8-Max Arena text: following instructions score (lmarena) |
| Arena text: business and management Arena 'text' leaderboard, category industry_business_and_management_and_financial_operations |
1,480 Arena score 95% range 1,472 to 1,489, 5,844 votes |
Top tier | effort: not stated, bradley-terry rating | Qwen3.8-Max lmarena · text/latest |
28 Oct 2026 | Source for Qwen3.8-Max Arena text: business and management score (lmarena) |
| Arena WebDev: building web apps Arena 'webdev' leaderboard (Code Arena), category overall |
1,672 Arena score 95% range 1,660 to 1,684, 3,406 votes |
Top tier | effort: not stated, bradley-terry rating | Qwen3.8-Max lmarena · webdev/latest |
28 Oct 2026 | Source for Qwen3.8-Max Arena WebDev: building web apps score (lmarena) |
| FrontierMath Tiers 1-3 (v2) FrontierMath Tiers 1-3 v2 (private set), benchmark released 2026-06-12 |
74.7% share of problems solved standard error ±2.6 points |
Strong | effort: xhigh, best score across scorers | qwen3.8-max_xhigh epoch · frontiermath_tiers_1_3_v2.csv |
4 Aug 2026 | Source for Qwen3.8-Max FrontierMath Tiers 1-3 (v2) score (epoch) · Run details for FrontierMath Tiers 1-3 (v2) |
| OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025, benchmark released 2024-12-19 |
99.4% share of problems solved standard error ±0.4 points |
Top tier | effort: xhigh, best score across scorers | qwen3.8-max_xhigh epoch · otis_mock_aime_2024_2025.csv |
4 Aug 2026 | Source for Qwen3.8-Max OTIS Mock AIME 2024-2025 score (epoch) No public run log in the data file. |
| GPQA Diamond GPQA Diamond, benchmark released 2023-11-20 |
92.7% share of questions correct standard error ±1.7 points |
Strong | effort: xhigh, best score across scorers | qwen3.8-max_xhigh epoch · gpqa_diamond.csv |
4 Aug 2026 | Source for Qwen3.8-Max GPQA Diamond score (epoch) No public run log in the data file. |
| SimpleQA Verified SimpleQA Verified, benchmark released 2025-09-09 |
45.8% share of questions correct standard error ±1.6 points |
Middle | effort: xhigh, best score across scorers | qwen3.8-max_xhigh epoch · simpleqa_verified.csv |
27 Aug 2026 | Source for Qwen3.8-Max SimpleQA Verified score (epoch) · Run details for SimpleQA Verified |
| SWE-bench Verified (Epoch runs) | Not enough data | |||||
| FrontierCode (reported by Cognition) | Not enough data | |||||
Missing evidence
Not enough data for these tests:
- SWE-bench Verified (Epoch runs)
- FrontierCode (reported by Cognition)
Missing results are not low scores and do not count against the model.
Where a source gives no public run log, we cannot show the run behind the score. See data sources.
Not sure this is the right one?
Tell us what you need help with and we'll narrow the options down.