DeepSeek-V2-Chat: benchmark scores, price and ranking
DeepSeek-V2-Chat from deepseek-ai ranks #404 of 743 LLMs on the LLMs Tiger aggregate benchmark score (47.8), based on 1 benchmark results captured up to 2026-10-13.
- Organisation
- deepseek-ai
- Type
- Open-weight / local model
- Open weights
- Yes
- Released
- 2024-05-06 (source: curated)
- Aggregate rank
- #404 of 743 (score 47.8)
Across the 1 benchmark where at least ten models are scored, DeepSeek-V2-Chat is in the top 10% on 0 and in the top half on 0.
DeepSeek-V2-Chat benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| MMLU-Pro | 54.8% | #69 of 136 | HuggingFace MMLU-Pro leaderboard | 2026-10-13 |
About DeepSeek-V2-Chat
DeepSeek-V2-Chat is an open-weight model that can run on your own hardware from deepseek-ai. It was released on 2024-05-06, according to the developer's announcement. No API price is listed for it yet. We hold 1 benchmark result for it across 1 area, and none is in the top tenth of its board.
Where DeepSeek-V2-Chat stands
On our aggregate score, DeepSeek-V2-Chat is #404 of 743 ranked models, around the middle of the table. Its score is 47.8; the scale is centred on 50, so it sits 2.2 points below the typical model. That beats 46% of the ranked field. Among the 7 ranked deepseek-ai models it is #3. The next one up is DeepSeek-V2.5 at 50.9. The next one down is DeepSeek-V2 at 46.6.
DeepSeek-V2-Chat, benchmark by benchmark
DeepSeek-V2-Chat lands below the middle on MMLU-Pro: 54.8%, #69 of 136, with 50% of the field scoring higher. Darwin-180B-RSI leads at 88.1%; the closest result above is Phi-3-medium-4k-instruct at 55.7%. It covers 57-subject knowledge + reasoning. The board median is 55.3%, so DeepSeek-V2-Chat is 0.4 points below it. Source: HuggingFace MMLU-Pro leaderboard, captured 2026-10-13.
Release history
The deepseek-ai model released before it was DeepSeek-V2 on 2024-05-06, which scores 46.6. The next one was DeepSeek-Coder-V2-Lite-Base on 2024-06-13, at 45.0.
Reading these numbers
The 1 score on this page comes from 1 source: HuggingFace MMLU-Pro leaderboard (1). No official leaderboard supplied any of them, so treat them as third-party measurements. With 1 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #401 Kimi K2 Thinking (47.8)
- #402 GPT-5.1 (Thinking, Medium) (47.8)
- #403 Tiny Recursion Model (TRM) (47.8)
- #405 Muse Code + Muse Spark 1.3 (47.8)
- #406 DeepSeek-V3.2 (47.7)
- #407 Grok 4 (Fast Reasoning) (47.7)
More from deepseek-ai
- DeepSeek R1 (0528) (#33, 53.6)
- DeepSeek-V2.5 (#162, 50.9)
- DeepSeek-V2 (#505, 46.6)
- DeepSeek-Coder-V2-Lite-Instruct (#546, 46.0)
- deepseek-math-7b-instruct (#604, 45.2)
- DeepSeek-Coder-V2-Lite-Base (#608, 45.0)
Compare it on the live leaderboard → · How the aggregate score works