Model comparison
Claude 2.1 vs DeepSeek-V3.1
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 25.2 on the Noometry Index.
Last verified . 4 shared benchmarks.
DeepSeekSummary
- They share 4 benchmarks with published results for both. Claude 2.1 scores higher in 0 categories and DeepSeek-V3.1 in 4 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek-V3.1 leads 38.9 to 10.2.
- The biggest single-benchmark swing is DTBench: 51% for Claude 2.1 and 82.7% for DeepSeek-V3.1.
- DeepSeek-V3.1 has downloadable open weights; the other is API-only.
Side by side
| Claude 2.1 | DeepSeek-V3.1 | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 25.2 | 42.8 |
| Released | 2023-11-21 | 2025-08-21 |
| Weights | Proprietary | Open |
| Context window | — | 164K |
| Max output | — | 8K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $0.95 |
| Results tracked | 7 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1 leads
Claude 2.1: 26.2 (#327), DeepSeek-V3.1: 40.3 (#144)
| Benchmark | Claude 2.1 | DeepSeek-V3.1 |
|---|---|---|
| WeirdML | 7.1% | 38.4% |
| LMArena Coding | — | 1417 |
Reasoning DeepSeek-V3.1 leads
Claude 2.1: 21.4 (#221), DeepSeek-V3.1: 27.9 (#110)
| Benchmark | Claude 2.1 | DeepSeek-V3.1 |
|---|---|---|
| DTBench | 51% | 82.7% |
| Epoch Capabilities Index | 119.27 | 139.92 |
| ForecastBench | 54.2 | 58 |
| SimpleBench | — | 40% |
| Kagi LLM Benchmark | — | 53.2% |
| LMArena Hard Prompts | — | 1417 |
| LMCA | — | 24.3% |
Math DeepSeek-V3.1 leads
Claude 2.1: 10.2 (#315), DeepSeek-V3.1: 38.9 (#122)
| Benchmark | Claude 2.1 | DeepSeek-V3.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.9% | — |
| LMArena Math | — | 1420 |
Knowledge DeepSeek-V3.1 leads
Claude 2.1: 15.4 (#292), DeepSeek-V3.1: 43.7 (#90)
| Benchmark | Claude 2.1 | DeepSeek-V3.1 |
|---|---|---|
| GPQA Diamond | 33% | — |
| Vectara Hallucination Rate | — | 5.5% |
| LMArena Expert | — | 1405 |
| MMLU | 73.5% | — |
Multilingual Not comparable
Claude 2.1: —, DeepSeek-V3.1: 51.6 (#106)
| Benchmark | Claude 2.1 | 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 2.1: —, DeepSeek-V3.1: 73.9 (#110)
| Benchmark | Claude 2.1 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Instruction Following | — | 1400 |
Long Context Not comparable
Claude 2.1: —, DeepSeek-V3.1: 36.3 (#232)
| Benchmark | Claude 2.1 | DeepSeek-V3.1 |
|---|---|---|
| Fiction.LiveBench | — | 52.8% |
| LMArena Longer Query | — | 1422 |
Writing & Preference Not comparable
Claude 2.1: —, DeepSeek-V3.1: 60.3 (#98)
| Benchmark | Claude 2.1 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Text | — | 1420 |
| LMArena Creative Writing | — | 1401 |
| EQ-Bench Creative Writing | — | 1436 |
| LMArena Multi-Turn | — | 1408 |
Frequently asked questions
Is Claude 2.1 better than DeepSeek-V3.1?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 25.2 on the Noometry Index.
Is Claude 2.1 or DeepSeek-V3.1 better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 26.2 in the Noometry coding category.
How many benchmarks do Claude 2.1 and DeepSeek-V3.1 share?
4 benchmarks have published results for both models. Claude 2.1 has 7 scored results on Noometry and DeepSeek-V3.1 has 27.