GPT-5.4
Made by OpenAI. Epoch AI does not state how it can be accessed. Appears solid at coding and writing among the models listed here.
In short
What it does well
- Coding: Strong (26th of 77 listed models)
- Writing: Strong (38th of 77 listed models)
Where it falls short
The four main ratings show no weakness. That is not evidence it has none.
What we don't know yet
- Hard maths problems: Not enough data
- School-competition maths: Not enough data
- Graduate-level science questions: Not enough data
- Short factual questions: Not enough data
- 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
- StrongHow often people preferred its answers to writing prompts. 1,439 on Arena's scale · Source for GPT-5.4 writing score (lmarena)
- Coding
- StrongHow often people preferred its answers to coding questions. 1,496 on Arena's scale · Source for GPT-5.4 coding score (lmarena)
- Maths
- Not enough dataNo result yet for hard maths problems.
- Knowledge
- Not enough dataNo result yet for short factual questions.
Evidence completeness: Partial evidence, 7 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
- Spanish: 41st of 70 on questions asked in Spanish, against 29th of 77 overall. Source for Spanish
Show every category we can see
| Category | Position | Arena score | Votes | Source |
|---|---|---|---|---|
| Expert-level prompts Kind of request | 19th of 77 | 1,506 95% range 1,498 to 1,514 | 6,361 | Source for Expert-level prompts |
| Hard prompts Kind of request | 24th of 77 | 1,485 95% range 1,481 to 1,489 | 42,125 | Source for Hard prompts |
| Long prompts Kind of request | 32nd of 77 | 1,473 95% range 1,468 to 1,478 | 28,430 | Source for Long prompts |
| Longer conversations Kind of request | 18th of 77 | 1,482 95% range 1,475 to 1,488 | 11,977 | Source for Longer conversations |
| Chinese Language | 27th of 76 | 1,519 95% range 1,509 to 1,529 | 4,119 | Source for Chinese |
| English prompts Language | 37th of 77 | 1,467 95% range 1,462 to 1,472 | 29,324 | Source for English prompts |
| French Language | 18th of 72 | 1,494 95% range 1,480 to 1,507 | 2,386 | Source for French |
| German Language | 20th of 55 | 1,472 95% range 1,454 to 1,491 | 1,093 | Source for German |
| Japanese Language | 13th of 54 | 1,483 95% range 1,459 to 1,508 | 681 | Source for Japanese |
| Korean Language | 14th of 54 | 1,448 95% range 1,429 to 1,468 | 1,114 | Source for Korean |
| Non-English prompts Language | 24th of 77 | 1,465 95% range 1,461 to 1,470 | 35,422 | Source for Non-English prompts |
| Russian Language | 21st of 77 | 1,480 95% range 1,472 to 1,488 | 7,199 | Source for Russian |
| Spanish Language | 41st of 70 | 1,453 95% range 1,439 to 1,467 | 2,066 | Source for Spanish |
| Entertainment, sport and media Topic | 34th of 77 | 1,438 95% range 1,432 to 1,444 | 14,150 | Source for Entertainment, sport and media |
| Law and government Topic | 20th of 76 | 1,486 95% range 1,477 to 1,495 | 5,227 | Source for Law and government |
| Medicine and health Topic | 41st of 75 | 1,458 95% range 1,449 to 1,467 | 4,827 | Source for Medicine and health |
| Science Topic | 32nd of 77 | 1,479 95% range 1,472 to 1,485 | 10,634 | Source for Science |
| Software and IT Topic | 28th of 77 | 1,491 95% range 1,486 to 1,496 | 25,478 | 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
- Homework and maths: not ranked, because there is not enough data for the main test.
- Research and checking facts: not ranked, because there is not enough data for the main test.
Where an alternative might be more suitable
- Writing and creating content: rated Strong on writing votes. Compare it with Claude Opus 5.5, the best-supported choice for this task.
- Coding and building things: rated Strong on coding votes. Compare it with Claude Opus 5.5, the best-supported choice for this task.
- Homework and maths: no result on hard maths problems, so we cannot say. See which models the evidence does support.
- Research and checking facts: no result on short factual questions, so we cannot say. See which models the evidence does support.
What it is
- Maker
- OpenAI
- Access, as listed by Epoch AI
- Not stated
- Licence, as listed by Arena
- Proprietary
- Last observation in our data
- 28 Oct 2026
- Data last changed
- 29 Oct 2026, 14:36 UTC
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. See models, apps and harnesses and how AI pricing works.
Price
Price not independently verified
See ChatGPT plans and pricing (checked 29 Oct 2026). Page opened and read in the browser. Which plan includes which model comes from its comparison table.
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) |
1,469 Arena score 95% range 1,465 to 1,473, 64,748 votes |
Strong | effort: High, bradley-terry rating | GPT-5.4 (High) lmarena · text/latest |
28 Oct 2026 | Source for GPT-5.4 Arena text: overall preference score (lmarena) |
| Arena text: creative writing Arena 'text' leaderboard, category creative_writing |
1,439 Arena score 95% range 1,432 to 1,445, 11,111 votes |
Strong | effort: High, bradley-terry rating | GPT-5.4 (High) lmarena · text/latest |
28 Oct 2026 | Source for GPT-5.4 Arena text: creative writing score (lmarena) |
| Arena text: coding Arena 'text' leaderboard, category coding |
1,496 Arena score 95% range 1,491 to 1,502, 17,603 votes |
Strong | effort: High, bradley-terry rating | GPT-5.4 (High) lmarena · text/latest |
28 Oct 2026 | Source for GPT-5.4 Arena text: coding score (lmarena) |
| Arena text: maths Arena 'text' leaderboard, category math |
1,488 Arena score 95% range 1,477 to 1,498, 3,378 votes |
Top tier | effort: High, bradley-terry rating | GPT-5.4 (High) lmarena · text/latest |
28 Oct 2026 | Source for GPT-5.4 Arena text: maths score (lmarena) |
| Arena text: following instructions Arena 'text' leaderboard, category instruction_following |
1,469 Arena score 95% range 1,464 to 1,475, 21,853 votes |
Strong | effort: High, bradley-terry rating | GPT-5.4 (High) lmarena · text/latest |
28 Oct 2026 | Source for GPT-5.4 Arena text: following instructions score (lmarena) |
| Arena text: business and management Arena 'text' leaderboard, category industry_business_and_management_and_financial_operations |
1,475 Arena score 95% range 1,468 to 1,481, 13,094 votes |
Top tier | effort: High, bradley-terry rating | GPT-5.4 (High) lmarena · text/latest |
28 Oct 2026 | Source for GPT-5.4 Arena text: business and management score (lmarena) |
| Arena WebDev: building web apps Arena 'webdev' leaderboard (Code Arena), category overall |
1,401 Arena score 95% range 1,391 to 1,411, 4,499 votes |
Behind the leaders | effort: not stated, bradley-terry rating | GPT-5.4 lmarena · webdev/latest |
28 Oct 2026 | Source for GPT-5.4 Arena WebDev: building web apps score (lmarena) |
| FrontierMath Tiers 1-3 (v2) | Not enough data | |||||
| OTIS Mock AIME 2024-2025 | Not enough data | |||||
| GPQA Diamond | Not enough data | |||||
| SimpleQA Verified | Not enough data | |||||
| SWE-bench Verified (Epoch runs) | Not enough data | |||||
| FrontierCode (reported by Cognition) | Not enough data | |||||
Missing evidence
Not enough data for these tests:
- FrontierMath Tiers 1-3 (v2)
- OTIS Mock AIME 2024-2025
- GPQA Diamond
- SimpleQA Verified
- 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.