Model comparison
Claude 3.7 Sonnet vs DeepSeek-V2.5 (Sep 2024)
Claude 3.7 Sonnet is the stronger model overall, scoring 39.4 to 37.0 on the Noometry Index.
Last verified . 18 shared benchmarks.
DeepSeekSummary
- They share 18 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 7 categories and DeepSeek-V2.5 (Sep 2024) in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in long context, where Claude 3.7 Sonnet leads 50.2 to 39.5.
- The biggest single-benchmark swing is Aider Polyglot: 64.9% for Claude 3.7 Sonnet and 17.8% for DeepSeek-V2.5 (Sep 2024).
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.
Side by side
| Claude 3.7 Sonnet | DeepSeek-V2.5 (Sep 2024) | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 39.4 | 37.0 |
| Released | 2025-02-24 | 2024-09-06 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 58 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 41.1 (#120), DeepSeek-V2.5 (Sep 2024): 31.3 (#274)
| Benchmark | Claude 3.7 Sonnet | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| Aider Polyglot | 64.9% | 17.8% |
| LMArena Coding | 1361 | 1310 |
| SWE-bench Verified | 61% | — |
| SWE-bench Verified (bash only) | 52.8% | — |
| GSO | 3.8% | — |
| BigCodeBench Instruct | — | 48.6% |
| LiveBench Coding | 74.5% | — |
| BigCodeBench Complete | — | 53.2% |
| CadEval | 54% | — |
| HumanEval+ | — | 83.5% |
| MBPP+ | — | 74.1% |
Agentic & Tool Use Not comparable
Claude 3.7 Sonnet: 33.9 (#46), DeepSeek-V2.5 (Sep 2024): —
| Benchmark | Claude 3.7 Sonnet | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| TheAgentCompany | 30.9% | — |
| Cybench | 20% | — |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
| METR Time Horizons | 60% | — |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
Claude 3.7 Sonnet: 18.7 (#252), DeepSeek-V2.5 (Sep 2024): 24.5 (#145)
| Benchmark | Claude 3.7 Sonnet | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Hard Prompts | 1333 | 1289 |
| ARC-AGI-2 | 0.9% | — |
| SimpleBench | 46.4% | — |
| ARC-AGI-1 | 28.6% | — |
| EnigmaEval | 4.2% | — |
| LiveBench Reasoning | 87.8% | — |
| LiveBench Data Analysis | 74% | — |
| Epoch Capabilities Index | 141.16 | — |
| ForecastBench | 61.8 | — |
| LiveBench | 76.1% | — |
Math Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 35.7 (#137), DeepSeek-V2.5 (Sep 2024): 33.3 (#177)
| Benchmark | Claude 3.7 Sonnet | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Math | 1336 | 1289 |
| OTIS Mock AIME 2024-2025 | 57.8% | — |
| Omni-MATH | 33% | — |
| LiveBench Math | 79% | — |
| MATH Level 5 | 91.2% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
Knowledge Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 39.7 (#117), DeepSeek-V2.5 (Sep 2024): 34.2 (#190)
| Benchmark | Claude 3.7 Sonnet | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Expert | 1321 | 1266 |
| GPQA Diamond | 79.7% | — |
| Humanity's Last Exam | 8% | — |
| MMLU-Pro | 78.4% | — |
| Confabulations | 14.7% | — |
| GPQA (HELM) | 60.8% | — |
Multimodal Not comparable
Claude 3.7 Sonnet: 33.1 (#90), DeepSeek-V2.5 (Sep 2024): —
| Benchmark | Claude 3.7 Sonnet | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Vision | 1169 | — |
| GeoBench | 68% | — |
| VPCT | 39% | — |
| SpatialViz-Bench | 33.9% | — |
Multilingual Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 44.3 (#174), DeepSeek-V2.5 (Sep 2024): 42.7 (#188)
| Benchmark | Claude 3.7 Sonnet | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Non-English | 1296 | 1273 |
| LMArena Chinese | 1299 | 1317 |
| LMArena French | 1302 | 1288 |
| LMArena German | 1301 | 1258 |
| LMArena Japanese | 1268 | 1228 |
| LMArena Korean | 1249 | 1209 |
| LMArena Russian | 1311 | 1289 |
| LMArena Spanish | 1299 | 1248 |
Instruction Following Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 73.1 (#120), DeepSeek-V2.5 (Sep 2024): 67.7 (#189)
| Benchmark | Claude 3.7 Sonnet | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Instruction Following | 1352 | 1280 |
| LiveBench Instruction Following | 81.3% | — |
| IFEval | 83.4% | — |
Long Context Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 50.2 (#10), DeepSeek-V2.5 (Sep 2024): 39.5 (#169)
| Benchmark | Claude 3.7 Sonnet | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Longer Query | 1373 | 1301 |
| Fiction.LiveBench | 83.3% | — |
Writing & Preference Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 54.6 (#145), DeepSeek-V2.5 (Sep 2024): 50.0 (#182)
| Benchmark | Claude 3.7 Sonnet | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Text | 1314 | 1294 |
| LMArena Creative Writing | 1332 | 1285 |
| LMArena Multi-Turn | 1339 | 1297 |
| Short-Story Creative Writing | 81.1% | — |
| EQ-Bench Creative Writing | 1412 | — |
| WildBench | 81.4% | — |
| LiveBench Language | 59.9% | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than DeepSeek-V2.5 (Sep 2024)?
Claude 3.7 Sonnet is the stronger model overall, scoring 39.4 to 37.0 on the Noometry Index.
Is Claude 3.7 Sonnet or DeepSeek-V2.5 (Sep 2024) better for coding?
Claude 3.7 Sonnet scores higher on coding benchmarks: 41.1 versus 31.3 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and DeepSeek-V2.5 (Sep 2024) share?
18 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and DeepSeek-V2.5 (Sep 2024) has 22.