Model comparison
Claude 2 vs DeepSeek-V3.1
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 25.0 on the Noometry Index.
Last verified . 2 shared benchmarks.
DeepSeekSummary
- They share 2 benchmarks with published results for both. Claude 2 scores higher in 0 categories and DeepSeek-V3.1 in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek-V3.1 leads 38.9 to 9.3.
- The biggest single-benchmark swing is DTBench: 51.9% for Claude 2 and 82.7% for DeepSeek-V3.1.
- DeepSeek-V3.1 has downloadable open weights; the other is API-only.
Side by side
| Claude 2 | DeepSeek-V3.1 | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 25.0 | 42.8 |
| Released | 2023-07-11 | 2025-08-21 |
| Weights | Proprietary | Open |
| Context window | — | 164K |
| Max output | — | 8K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $0.95 |
| Results tracked | 8 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Claude 2: —, DeepSeek-V3.1: 40.3 (#144)
| Benchmark | Claude 2 | DeepSeek-V3.1 |
|---|---|---|
| WeirdML | — | 38.4% |
| LMArena Coding | — | 1417 |
| HumanEval+ | 61.6% | — |
Reasoning DeepSeek-V3.1 leads
Claude 2: 21.7 (#216), DeepSeek-V3.1: 27.9 (#110)
| Benchmark | Claude 2 | DeepSeek-V3.1 |
|---|---|---|
| DTBench | 51.9% | 82.7% |
| Epoch Capabilities Index | 120.13 | 139.92 |
| SimpleBench | — | 40% |
| Kagi LLM Benchmark | — | 53.2% |
| LMArena Hard Prompts | — | 1417 |
| LMCA | — | 24.3% |
| ForecastBench | — | 58 |
Math DeepSeek-V3.1 leads
Claude 2: 9.3 (#320), DeepSeek-V3.1: 38.9 (#122)
| Benchmark | Claude 2 | DeepSeek-V3.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.5% | — |
| LMArena Math | — | 1420 |
| MATH Level 5 | 11.7% | — |
Knowledge DeepSeek-V3.1 leads
Claude 2: 16.9 (#287), DeepSeek-V3.1: 43.7 (#90)
| Benchmark | Claude 2 | DeepSeek-V3.1 |
|---|---|---|
| GPQA Diamond | 34.7% | — |
| Vectara Hallucination Rate | — | 5.5% |
| LMArena Expert | — | 1405 |
| MMLU | 78.5% | — |
| TriviaQA | 87.5% | — |
Multilingual Not comparable
Claude 2: —, DeepSeek-V3.1: 51.6 (#106)
| Benchmark | Claude 2 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Non-English | — | 1400 |
| LMArena Chinese | — | 1469 |
| LMArena French | — | 1447 |
| LMArena German | — | 1411 |
| LMArena Japanese | — | 1378 |
| LMArena Korean | — | 1337 |
| LMArena Russian | — | 1405 |
| LMArena Spanish | — | 1431 |
Instruction Following Not comparable
Claude 2: —, DeepSeek-V3.1: 73.9 (#110)
| Benchmark | Claude 2 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Instruction Following | — | 1400 |
Long Context Not comparable
Claude 2: —, DeepSeek-V3.1: 36.3 (#232)
| Benchmark | Claude 2 | DeepSeek-V3.1 |
|---|---|---|
| Fiction.LiveBench | — | 52.8% |
| LMArena Longer Query | — | 1422 |
Writing & Preference Not comparable
Claude 2: —, DeepSeek-V3.1: 60.3 (#98)
| Benchmark | Claude 2 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Text | — | 1420 |
| LMArena Creative Writing | — | 1401 |
| EQ-Bench Creative Writing | — | 1436 |
| LMArena Multi-Turn | — | 1408 |
Frequently asked questions
Is Claude 2 better than DeepSeek-V3.1?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 25.0 on the Noometry Index.
How many benchmarks do Claude 2 and DeepSeek-V3.1 share?
2 benchmarks have published results for both models. Claude 2 has 8 scored results on Noometry and DeepSeek-V3.1 has 27.