Model comparison
Claude 3 Opus vs DeepSeek-V3.1-Terminus
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 29.3 on the Noometry Index.
Last verified . 12 shared benchmarks.
DeepSeekSummary
- They share 12 benchmarks with published results for both. Claude 3 Opus scores higher in 0 categories and DeepSeek-V3.1-Terminus in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek-V3.1-Terminus leads 35.7 to 13.9.
- The biggest single-benchmark swing is DTBench: 61.6% for Claude 3 Opus and 81.3% for DeepSeek-V3.1-Terminus.
- DeepSeek-V3.1-Terminus has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Opus | DeepSeek-V3.1-Terminus | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 29.3 | 42.6 |
| Released | 2024-02-29 | 2025-09-22 |
| Weights | Proprietary | Open |
| Context window | — | 164K |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.30 |
| Output $ / M tokens | — | $1 |
| Results tracked | 46 | 16 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1-Terminus leads
Claude 3 Opus: 32.4 (#261), DeepSeek-V3.1-Terminus: 41.5 (#114)
| Benchmark | Claude 3 Opus | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Coding | 1264 | 1426 |
| SciCode | — | 40.6% |
| WeirdML | 19.2% | — |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 38.6% | — |
| BigCodeBench Complete | 57.4% | — |
| ALE-Bench | — | 745.17 |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Not comparable
Claude 3 Opus: 24.3 (#110), DeepSeek-V3.1-Terminus: —
| Benchmark | Claude 3 Opus | DeepSeek-V3.1-Terminus |
|---|---|---|
| Cybench | 10% | — |
| METR Time Horizons | 29.5% | — |
Reasoning DeepSeek-V3.1-Terminus leads
Claude 3 Opus: 14.2 (#318), DeepSeek-V3.1-Terminus: 25.7 (#126)
| Benchmark | Claude 3 Opus | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Hard Prompts | 1246 | 1427 |
| DTBench | 61.6% | 81.3% |
| LMCA | 17% | 28.6% |
| SimpleBench | 23.5% | — |
| Kagi LLM Benchmark | — | 57.4% |
| CritPt | — | 1.7% |
| Chess Puzzles | 5% | — |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 40.6% | — |
| LiveBench Data Analysis | 57.9% | — |
| Epoch Capabilities Index | 126.91 | — |
| ForecastBench | 58.4 | — |
| LiveBench | 49.2% | — |
| WinoGrande | 88.5% | — |
Math DeepSeek-V3.1-Terminus leads
Claude 3 Opus: 13.9 (#293), DeepSeek-V3.1-Terminus: 35.7 (#138)
| Benchmark | Claude 3 Opus | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Math | 1273 | 1402 |
| OTIS Mock AIME 2024-2025 | 4.7% | — |
| LiveBench Math | 43.6% | — |
| MATH Level 5 | 37.5% | — |
Knowledge Not comparable
Claude 3 Opus: 24.5 (#262), DeepSeek-V3.1-Terminus: —
| Benchmark | Claude 3 Opus | DeepSeek-V3.1-Terminus |
|---|---|---|
| GPQA Diamond | 47.2% | — |
| SimpleQA Verified | 12.6% | — |
| Confabulations | 22.7% | — |
| LMArena Expert | 1223 | — |
| MMLU | 84.6% | — |
Multimodal Not comparable
Claude 3 Opus: 26.6 (#112), DeepSeek-V3.1-Terminus: —
| Benchmark | Claude 3 Opus | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Vision | 1023 | — |
Multilingual DeepSeek-V3.1-Terminus leads
Claude 3 Opus: 41.6 (#202), DeepSeek-V3.1-Terminus: 52.3 (#87)
| Benchmark | Claude 3 Opus | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Non-English | 1258 | 1407 |
| LMArena Russian | 1280 | 1434 |
| LMArena Chinese | 1248 | — |
| LMArena French | 1274 | — |
| LMArena German | 1258 | — |
| LMArena Japanese | 1204 | — |
| LMArena Korean | 1187 | — |
| LMArena Spanish | 1246 | — |
Instruction Following DeepSeek-V3.1-Terminus leads
Claude 3 Opus: 64.3 (#223), DeepSeek-V3.1-Terminus: 74.4 (#96)
| Benchmark | Claude 3 Opus | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Instruction Following | 1248 | 1406 |
| LiveBench Instruction Following | 63.9% | — |
Long Context DeepSeek-V3.1-Terminus leads
Claude 3 Opus: 38.2 (#197), DeepSeek-V3.1-Terminus: 43.5 (#88)
| Benchmark | Claude 3 Opus | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Longer Query | 1259 | 1426 |
Writing & Preference DeepSeek-V3.1-Terminus leads
Claude 3 Opus: 47.4 (#208), DeepSeek-V3.1-Terminus: 61.3 (#87)
| Benchmark | Claude 3 Opus | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Text | 1262 | 1420 |
| LMArena Creative Writing | 1235 | 1404 |
| LMArena Multi-Turn | 1275 | 1411 |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than DeepSeek-V3.1-Terminus?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 29.3 on the Noometry Index.
Is Claude 3 Opus or DeepSeek-V3.1-Terminus better for coding?
DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 41.5 versus 32.4 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and DeepSeek-V3.1-Terminus share?
12 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and DeepSeek-V3.1-Terminus has 16.