Model comparison
Claude Opus 4 vs DeepSeek-V3.1
Claude Opus 4 and DeepSeek-V3.1 score almost the same on the Noometry Index (43.1 vs 42.8), so choose on price, context window or the category you care about most.
Last verified . 27 shared benchmarks.
DeepSeekSummary
- They share 27 benchmarks with published results for both. Claude Opus 4 scores higher in 6 categories and DeepSeek-V3.1 in 2 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in coding, where Claude Opus 4 leads 47.2 to 40.3.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 74.3% for Claude Opus 4 and 53.2% for DeepSeek-V3.1.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $15 / $75 for Claude Opus 4.
- Claude Opus 4 accepts more context: 200K tokens versus 164K.
- DeepSeek-V3.1 has downloadable open weights; the other is API-only.
Side by side
| Claude Opus 4 | DeepSeek-V3.1 | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 43.1 | 42.8 |
| Released | 2025-05-22 | 2025-08-21 |
| Weights | Proprietary | Open |
| Context window | 200K | 164K |
| Max output | 32K | 8K |
| Input $ / M tokens | $15 | $0.25 |
| Output $ / M tokens | $75 | $0.95 |
| Results tracked | 56 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4 leads
Claude Opus 4: 47.2 (#62), DeepSeek-V3.1: 40.3 (#144)
| Benchmark | Claude Opus 4 | DeepSeek-V3.1 |
|---|---|---|
| WeirdML | 43.7% | 38.4% |
| LMArena Coding | 1442 | 1417 |
| SWE-bench Verified | 70.7% | — |
| SWE-bench Verified (bash only) | 67.6% | — |
| Aider Polyglot | 72% | — |
| GSO | 6.9% | — |
| AlgoTune | 1.33 | — |
Agentic & Tool Use Not comparable
Claude Opus 4: 34.8 (#42), DeepSeek-V3.1: —
| Benchmark | Claude Opus 4 | DeepSeek-V3.1 |
|---|---|---|
| Cybench | 38% | — |
| DeepResearch Bench | 46.8% | — |
| LMArena Search | 1127 | — |
| METR Time Horizons | 63.9% | — |
Reasoning Too close to call
Claude Opus 4: 27.3 (#121), DeepSeek-V3.1: 27.9 (#110)
| Benchmark | Claude Opus 4 | DeepSeek-V3.1 |
|---|---|---|
| SimpleBench | 58.8% | 40% |
| Kagi LLM Benchmark | 74.3% | 53.2% |
| LMArena Hard Prompts | 1399 | 1417 |
| DTBench | 81.6% | 82.7% |
| LMCA | 37.4% | 24.3% |
| Epoch Capabilities Index | 142.67 | 139.92 |
| ForecastBench | 61.1 | 58 |
| ARC-AGI-2 | 8.6% | — |
| ARC-AGI-1 | 35.7% | — |
| CritPt | 0.3% | — |
| EnigmaEval | 5.6% | — |
Math Claude Opus 4 leads
Claude Opus 4: 42.0 (#86), DeepSeek-V3.1: 38.9 (#122)
| Benchmark | Claude Opus 4 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Math | 1390 | 1420 |
| OTIS Mock AIME 2024-2025 | 64.4% | — |
| Omni-MATH | 61.6% | — |
| MATH Level 5 | 85% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Too close to call
Claude Opus 4: 44.0 (#88), DeepSeek-V3.1: 43.7 (#90)
| Benchmark | Claude Opus 4 | DeepSeek-V3.1 |
|---|---|---|
| Vectara Hallucination Rate | 12% | 5.5% |
| LMArena Expert | 1386 | 1405 |
| GPQA Diamond | 76.3% | — |
| Humanity's Last Exam | 10.7% | — |
| MMLU-Pro | 87.5% | — |
| Confabulations | 15.9% | — |
| GPQA (HELM) | 70.8% | — |
Multimodal Not comparable
Claude Opus 4: 31.5 (#106), DeepSeek-V3.1: —
| Benchmark | Claude Opus 4 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Vision | 1192 | — |
| GeoBench | 49% | — |
| VPCT | 38% | — |
Multilingual DeepSeek-V3.1 leads
Claude Opus 4: 48.8 (#138), DeepSeek-V3.1: 51.6 (#106)
| Benchmark | Claude Opus 4 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Non-English | 1362 | 1400 |
| LMArena Chinese | 1386 | 1469 |
| LMArena French | 1372 | 1447 |
| LMArena German | 1391 | 1411 |
| LMArena Japanese | 1331 | 1378 |
| LMArena Korean | 1321 | 1337 |
| LMArena Russian | 1392 | 1405 |
| LMArena Spanish | 1389 | 1431 |
Instruction Following Claude Opus 4 leads
Claude Opus 4: 77.1 (#28), DeepSeek-V3.1: 73.9 (#110)
| Benchmark | Claude Opus 4 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Instruction Following | 1406 | 1400 |
| IFEval | 91.8% | — |
Long Context Claude Opus 4 leads
Claude Opus 4: 39.6 (#172), DeepSeek-V3.1: 36.3 (#232)
| Benchmark | Claude Opus 4 | DeepSeek-V3.1 |
|---|---|---|
| Fiction.LiveBench | 61.1% | 52.8% |
| LMArena Longer Query | 1422 | 1422 |
Writing & Preference Too close to call
Claude Opus 4: 61.2 (#89), DeepSeek-V3.1: 60.3 (#98)
| Benchmark | Claude Opus 4 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Text | 1377 | 1420 |
| LMArena Creative Writing | 1387 | 1401 |
| EQ-Bench Creative Writing | 1580 | 1436 |
| LMArena Multi-Turn | 1396 | 1408 |
| Short-Story Creative Writing | 83.6% | — |
| WildBench | 85.2% | — |
Frequently asked questions
Is Claude Opus 4 better than DeepSeek-V3.1?
Claude Opus 4 and DeepSeek-V3.1 score almost the same on the Noometry Index (43.1 vs 42.8), so choose on price, context window or the category you care about most.
Which is cheaper, Claude Opus 4 or DeepSeek-V3.1?
DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; Claude Opus 4 lists at $15 and $75.
Is Claude Opus 4 or DeepSeek-V3.1 better for coding?
Claude Opus 4 scores higher on coding benchmarks: 47.2 versus 40.3 in the Noometry coding category.
Which has the bigger context window?
Claude Opus 4 does, with 200K tokens against 164K.
How many benchmarks do Claude Opus 4 and DeepSeek-V3.1 share?
27 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and DeepSeek-V3.1 has 27.