Model comparison
Claude 3 Opus vs DeepSeek V4.1 Flash
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 29.5 on the Noometry Index.
Last verified . 23 shared benchmarks.
DeepSeekSummary
- They share 23 benchmarks with published results for both. Claude 3 Opus scores higher in 0 categories and DeepSeek V4.1 Flash in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4.1 Flash leads 66.7 to 14.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.7% for Claude 3 Opus and 98.3% for DeepSeek V4.1 Flash.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Opus | DeepSeek V4.1 Flash | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 29.5 | 52.8 |
| Released | 2024-02-29 | 2026-09-09 |
| Weights | Proprietary | Open |
| Context window | — | 1M |
| Max output | — | 393K |
| Input $ / M tokens | — | $0.15 |
| Output $ / M tokens | — | $0.60 |
| Results tracked | 46 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4.1 Flash leads
Claude 3 Opus: 32.9 (#267), DeepSeek V4.1 Flash: 52.9 (#32)
| Benchmark | Claude 3 Opus | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Coding | 1264 | 1506 |
| LMArena WebDev | — | 1619 |
| SciCode | — | 51.9% |
| WeirdML | 19.2% | — |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 38.6% | — |
| BigCodeBench Complete | 57.4% | — |
| ALE-Bench | — | 1,092 |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use DeepSeek V4.1 Flash leads
Claude 3 Opus: 24.6 (#116), DeepSeek V4.1 Flash: 31.2 (#69)
| Benchmark | Claude 3 Opus | DeepSeek V4.1 Flash |
|---|---|---|
| APEX-Agents | — | 39.5% |
| Cybench | 10% | — |
| GDP.pdf | — | 19.8% |
| METR Time Horizons | 29.5% | — |
Reasoning DeepSeek V4.1 Flash leads
Claude 3 Opus: 14.6 (#324), DeepSeek V4.1 Flash: 50.2 (#36)
| Benchmark | Claude 3 Opus | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Hard Prompts | 1245 | 1483 |
| DTBench | 61.6% | 89.9% |
| LMCA | 17% | 47% |
| Epoch Capabilities Index | 126.91 | 154.9 |
| SimpleBench | 23.5% | — |
| NYT Connections (extended) | — | 89.6% |
| CritPt | — | 14.3% |
| Chess Puzzles | 5% | — |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 40.6% | — |
| Mystery Game Puzzles | — | 43% |
| LiveBench Data Analysis | 57.9% | — |
| Surface Evolver Bench | — | 46.3% |
| ForecastBench | 58.4 | — |
| LiveBench | 49.2% | — |
| WinoGrande | 88.5% | — |
Math DeepSeek V4.1 Flash leads
Claude 3 Opus: 14.8 (#299), DeepSeek V4.1 Flash: 66.7 (#25)
| Benchmark | Claude 3 Opus | DeepSeek V4.1 Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 98.3% |
| LMArena Math | 1273 | 1477 |
| FrontierMath (Tiers 1-3) | — | 67.4% |
| FrontierMath Tier 4 | — | 26.8% |
| ProofBench | — | 54% |
| LiveBench Math | 43.6% | — |
| MATH Level 5 | 37.5% | — |
Knowledge DeepSeek V4.1 Flash leads
Claude 3 Opus: 24.5 (#267), DeepSeek V4.1 Flash: 57.9 (#38)
| Benchmark | Claude 3 Opus | DeepSeek V4.1 Flash |
|---|---|---|
| GPQA Diamond | 47.2% | 89.8% |
| LMArena Expert | 1223 | 1506 |
| SimpleQA Verified | 12.6% | — |
| Confabulations | 22.7% | — |
| MMLU | 84.6% | — |
Multimodal DeepSeek V4.1 Flash leads
Claude 3 Opus: 27.1 (#116), DeepSeek V4.1 Flash: 39.1 (#61)
| Benchmark | Claude 3 Opus | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Vision | 1023 | 1277 |
| Furniture Assembly | — | 34.2% |
Multilingual DeepSeek V4.1 Flash leads
Claude 3 Opus: 41.4 (#207), DeepSeek V4.1 Flash: 55.0 (#35)
| Benchmark | Claude 3 Opus | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Non-English | 1258 | 1448 |
| LMArena Chinese | 1248 | 1497 |
| LMArena French | 1275 | 1452 |
| LMArena German | 1258 | 1484 |
| LMArena Japanese | 1204 | 1412 |
| LMArena Korean | 1187 | 1452 |
| LMArena Russian | 1280 | 1471 |
| LMArena Spanish | 1246 | 1459 |
Instruction Following DeepSeek V4.1 Flash leads
Claude 3 Opus: 64.1 (#228), DeepSeek V4.1 Flash: 77.3 (#26)
| Benchmark | Claude 3 Opus | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Instruction Following | 1248 | 1474 |
| LiveBench Instruction Following | 63.9% | — |
Long Context DeepSeek V4.1 Flash leads
Claude 3 Opus: 38.2 (#202), DeepSeek V4.1 Flash: 45.2 (#47)
| Benchmark | Claude 3 Opus | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Longer Query | 1259 | 1475 |
Writing & Preference DeepSeek V4.1 Flash leads
Claude 3 Opus: 47.2 (#213), DeepSeek V4.1 Flash: 65.4 (#48)
| Benchmark | Claude 3 Opus | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Text | 1262 | 1462 |
| LMArena Creative Writing | 1235 | 1435 |
| LMArena Multi-Turn | 1275 | 1457 |
| EQ-Bench Creative Writing | — | 1540 |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than DeepSeek V4.1 Flash?
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 29.5 on the Noometry Index.
Is Claude 3 Opus or DeepSeek V4.1 Flash better for coding?
DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 32.9 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and DeepSeek V4.1 Flash share?
23 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and DeepSeek V4.1 Flash has 37.