DeepSeek-V4-Pro Preview
Made by DeepSeek. Epoch AI does not state how it can be accessed. Appears solid at writing among the models listed here.
In short
What it does well
- Writing: Strong (35th of 77 listed models)
Where it falls short
- Coding: Behind the leaders (67th of 77 listed models)
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. 1,441 on Arena's scale · Source for DeepSeek-V4-Pro Preview writing score (lmarena)
- Coding
- Behind the leadersHow often people preferred its answers to coding questions. 1,453 on Arena's scale · Source for DeepSeek-V4-Pro Preview 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
- Medicine and health: 34th of 75 on medicine and health questions, against 55th of 77 overall. Source for Medicine and health
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 | 62nd of 77 | 1,449 95% range 1,441 to 1,458 | 5,546 | Source for Expert-level prompts |
| Hard prompts Kind of request | 56th of 77 | 1,452 95% range 1,448 to 1,457 | 36,907 | Source for Hard prompts |
| Long prompts Kind of request | 53rd of 77 | 1,446 95% range 1,440 to 1,451 | 24,713 | Source for Long prompts |
| Longer conversations Kind of request | 57th of 77 | 1,446 95% range 1,438 to 1,453 | 9,688 | Source for Longer conversations |
| Chinese Language | 55th of 76 | 1,486 95% range 1,475 to 1,497 | 3,325 | Source for Chinese |
| English prompts Language | 48th of 77 | 1,455 95% range 1,450 to 1,460 | 23,986 | Source for English prompts |
| French Language | 59th of 72 | 1,451 95% range 1,437 to 1,466 | 2,065 | Source for French |
| German Language | 39th of 55 | 1,448 95% range 1,428 to 1,468 | 922 | Source for German |
| Japanese Language | 39th of 54 | 1,413 95% range 1,390 to 1,437 | 691 | Source for Japanese |
| Korean Language | 29th of 54 | 1,424 95% range 1,405 to 1,444 | 986 | Source for Korean |
| Non-English prompts Language | 56th of 77 | 1,431 95% range 1,426 to 1,436 | 30,880 | Source for Non-English prompts |
| Russian Language | 47th of 77 | 1,449 95% range 1,441 to 1,458 | 6,144 | Source for Russian |
| Spanish Language | 46th of 70 | 1,448 95% range 1,433 to 1,463 | 1,680 | Source for Spanish |
| Entertainment, sport and media Topic | 45th of 77 | 1,427 95% range 1,420 to 1,434 | 12,385 | Source for Entertainment, sport and media |
| Law and government Topic | 43rd of 76 | 1,459 95% range 1,450 to 1,469 | 4,331 | Source for Law and government |
| Medicine and health Topic | 34th of 75 | 1,468 95% range 1,458 to 1,478 | 3,849 | Source for Medicine and health |
| Science Topic | 44th of 77 | 1,463 95% range 1,456 to 1,470 | 8,793 | Source for Science |
| Software and IT Topic | 64th of 77 | 1,457 95% range 1,451 to 1,462 | 22,413 | 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 Behind the leaders 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.
- Business emails and everyday work: rated Behind the leaders on business-task votes. Compare it with Gemini 4 Argon, the best-supported choice for this task.
- 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
- DeepSeek
- Access, as listed by Epoch AI
- Not stated
- Licence, as listed by Arena
- MIT
- 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
Open weights. Arena lists its licence as MIT. In plain terms, the model's files are published under a licence that lets others run it, so it can be offered by more than one company, sometimes at lower cost. That does not guarantee that any particular app is free. Open licences differ, and some restrict commercial use, so check the licence terms before building a business on it.
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 DeepSeek API models and pricing (checked 29 Oct 2026). Page opened and read. It gives developer (API) prices per 1M tokens, shown on model pages. The chat app (chat.deepseek.com) goes straight to a sign-in page and we did not sign in, so we make no statement about whether it is free.
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) |
1,445 Arena score 95% range 1,441 to 1,449, 54,870 votes |
Middle | effort: High, bradley-terry rating | DeepSeek-V4-Pro Preview (High) lmarena · text/latest |
28 Oct 2026 | Source for DeepSeek-V4-Pro Preview Arena text: overall preference score (lmarena) |
| Arena text: creative writing Arena 'text' leaderboard, category creative_writing |
1,441 Arena score 95% range 1,434 to 1,448, 9,369 votes |
Strong | effort: High, bradley-terry rating | DeepSeek-V4-Pro Preview (High) lmarena · text/latest |
28 Oct 2026 | Source for DeepSeek-V4-Pro Preview Arena text: creative writing score (lmarena) |
| Arena text: coding Arena 'text' leaderboard, category coding |
1,453 Arena score 95% range 1,447 to 1,459, 16,003 votes |
Behind the leaders | effort: High, bradley-terry rating | DeepSeek-V4-Pro Preview (High) lmarena · text/latest |
28 Oct 2026 | Source for DeepSeek-V4-Pro Preview Arena text: coding score (lmarena) |
| Arena text: maths Arena 'text' leaderboard, category math |
1,454 Arena score 95% range 1,442 to 1,466, 2,714 votes |
Middle | effort: High, bradley-terry rating | DeepSeek-V4-Pro Preview (High) lmarena · text/latest |
28 Oct 2026 | Source for DeepSeek-V4-Pro Preview Arena text: maths score (lmarena) |
| Arena text: following instructions Arena 'text' leaderboard, category instruction_following |
1,436 Arena score 95% range 1,430 to 1,441, 19,046 votes |
Middle | effort: High, bradley-terry rating | DeepSeek-V4-Pro Preview (High) lmarena · text/latest |
28 Oct 2026 | Source for DeepSeek-V4-Pro Preview Arena text: following instructions score (lmarena) |
| Arena text: business and management Arena 'text' leaderboard, category industry_business_and_management_and_financial_operations |
1,428 Arena score 95% range 1,421 to 1,434, 11,029 votes |
Behind the leaders | effort: High, bradley-terry rating | DeepSeek-V4-Pro Preview (High) lmarena · text/latest |
28 Oct 2026 | Source for DeepSeek-V4-Pro Preview Arena text: business and management score (lmarena) |
| Arena WebDev: building web apps Arena 'webdev' leaderboard (Code Arena), category overall |
1,464 Arena score 95% range 1,458 to 1,471, 13,087 votes |
Middle | effort: High, bradley-terry rating | DeepSeek-V4-Pro Preview (High) lmarena · webdev/latest |
28 Oct 2026 | Source for DeepSeek-V4-Pro Preview 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.