DeepSeek V3 0324: benchmark scores, price and ranking
DeepSeek V3 0324 from DeepSeek ranks #169 of 743 LLMs on the LLMs Tiger aggregate benchmark score (50.8), based on 3 benchmark results captured up to 2026-10-13.
- Organisation
- DeepSeek
- Type
- Open-weight / local model
- Open weights
- Yes
- Released
- 2025-03-24 (source: curated)
- Cheapest API price
- $0.200 in / $0.600 out per million tokens (via lambda_ai, observed 2026-10-13T20:47Z)
- Aggregate rank
- #169 of 743 (score 50.8)
Relative to other models, DeepSeek V3 0324 is strongest on MMLU-Pro (#30 of 136) and weakest on SWE-bench Verified (#51 of 73).
Across the 3 benchmarks where at least ten models are scored, DeepSeek V3 0324 is in the top 10% on 0 and in the top half on 2.
Its cheapest listed input price, $0.200 per million tokens, is lower than that of 67% of the 427 models with a listed price.
DeepSeek V3 0324 benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| MMLU-Pro | 81.2% | #30 of 136 | HuggingFace MMLU-Pro leaderboard | 2026-10-13 |
| Aider Polyglot | 55.1% | #15 of 49 | Aider polyglot leaderboard | 2026-10-13 |
| SWE-bench Verified | 42.0% | #51 of 73 | SWE-bench Verified leaderboard | 2026-10-13 |
About DeepSeek V3 0324
DeepSeek V3 0324 is an open-weight model that can run on your own hardware from DeepSeek. It was released on 2025-03-24, according to the developer's announcement. The cheapest API listing we track charges $0.200 per million input tokens and $0.600 per million output tokens, through lambda_ai. We hold 3 benchmark results for it across 2 areas, and none is in the top tenth of its board.
Where DeepSeek V3 0324 stands
On our aggregate score, DeepSeek V3 0324 is #169 of 743 ranked models, in the top quarter. Its score is 50.8; the scale is centred on 50, so it sits 0.8 points above the typical model. That beats 77% of the ranked field. Among the 33 ranked DeepSeek models it is #8. The next one up is DeepSeek Chat V2 0628 at 51.4. The next one down is DeepSeek Chat V3 (prev) at 50.5.
DeepSeek V3 0324, benchmark by benchmark
MMLU-Pro puts DeepSeek V3 0324 at 81.2%, #30 of 136. That beats 78% of the models tested, but 6.9 points short of Darwin-180B-RSI (88.1%). The results either side are GLM-4.5-Air (81.4%) and MiniMax-M1-40k (81.1%). It covers 57-subject knowledge + reasoning. The board median is 55.3%, so DeepSeek V3 0324 is 25.9 points above it. Source: HuggingFace MMLU-Pro leaderboard, captured 2026-10-13.
On Aider Polyglot, DeepSeek V3 0324 reaches 55.1%. It ranks #15 of 49, ahead of 69% of the field, and 28.0 points behind gemini-2.5-pro-preview-06-05 at 83.1%. The results either side are DeepSeek-R1 (56.9%) and gemini-2.5-flash-preview-05-20 (55.1%). It covers edits across 6 languages (225 exercises). The board median is 44.9%, so DeepSeek V3 0324 is 10.2 points above it. Source: Aider polyglot leaderboard, captured 2026-10-13.
DeepSeek V3 0324 lands below the middle on SWE-bench Verified: 42.0%, #51 of 73, with 68% of the field scoring higher. Claude Opus 4.7 (max) leads at 83.5%; the closest result above is DeepSWE-Preview at 42.2%. It covers real GitHub issues resolved. The board median is 56.4%, so DeepSeek V3 0324 is 14.4 points below it. Source: SWE-bench Verified leaderboard, captured 2026-10-13.
Strengths and weaknesses
Software engineering: uneven, ahead of 50% of the field on average across 2 benchmarks. Its best board here is Aider Polyglot (#15 of 49) and its weakest is SWE-bench Verified (#51 of 73).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so DeepSeek V3 0324 costs less than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4 Flash 0731 (max) ($0.016, 53.1), GLM-5.3-Flash (max) ($0.026, 49.6). Similar scores cost more on Grok 3 mini ($0.250) and Qwen3.5 27B ($0.260).
Release history
The DeepSeek model released before it was DeepSeek-R1-Distill-Qwen-32B on 2025-01-20, which scores 47.9. The next one was DeepSeek-V3 (Mar 2025) on 2025-03-24, at 44.1.
Reading these numbers
The 3 scores on this page come from 3 sources: Aider polyglot leaderboard (1), HuggingFace MMLU-Pro leaderboard (1), SWE-bench Verified leaderboard (1). 1 of them come from an official leaderboard and 2 from datasets or third-party runs. With 3 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #166 gpt-4-0314 (50.8)
- #167 SWE-1.7 (50.8)
- #168 claude-3-5-haiku-20241022 (50.8)
- #170 o3-mini_1.0 (50.7)
- #171 claude-opus-4-1-20250805_27K (50.7)
- #172 GPT-5.1 Codex (50.7)
More from DeepSeek
- DeepSeek V4 Pro 0813 (max) (#23, 54.3)
- DeepSeek V4 Flash 0731 (max) (#42, 53.1)
- DeepSeek-V3.2-Exp (high) (#67, 52.4)
- deepseek-v4.1-flash-max (#75, 52.2)
- DeepSeek-V2.5-1210 (#101, 51.7)
- DeepSeek v4 Pro (max) (#108, 51.6)
Compare it on the live leaderboard → · How the aggregate score works