Model comparison
Claude Sonnet 4 vs DeepSeek V4.1 Flash
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 40.8 on the Noometry Index.
Last verified . 27 shared benchmarks.
DeepSeekSummary
- They share 27 benchmarks with published results for both. Claude Sonnet 4 scores higher in 1 category and DeepSeek V4.1 Flash in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4.1 Flash leads 50.2 to 22.9.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 71.1% for Claude Sonnet 4 and 98.3% for DeepSeek V4.1 Flash.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- DeepSeek V4.1 Flash accepts more context: 1M tokens versus 200K.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4 | DeepSeek V4.1 Flash | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 40.8 | 52.8 |
| Released | 2025-05-22 | 2026-09-09 |
| Weights | Proprietary | Open |
| Context window | 200K | 1M |
| Max output | 64K | 393K |
| Input $ / M tokens | $3 | $0.15 |
| Output $ / M tokens | $15 | $0.60 |
| Results tracked | 58 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4.1 Flash leads
Claude Sonnet 4: 43.5 (#88), DeepSeek V4.1 Flash: 52.9 (#32)
| Benchmark | Claude Sonnet 4 | DeepSeek V4.1 Flash |
|---|---|---|
| SciCode | 40% | 51.9% |
| LMArena Coding | 1414 | 1506 |
| ALE-Bench | 655.35 | 1,092 |
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| LMArena WebDev | — | 1619 |
| GSO | 4.9% | — |
| WeirdML | 46.1% | — |
Agentic & Tool Use Claude Sonnet 4 leads
Claude Sonnet 4: 38.5 (#31), DeepSeek V4.1 Flash: 31.2 (#69)
| Benchmark | Claude Sonnet 4 | DeepSeek V4.1 Flash |
|---|---|---|
| APEX-Agents | — | 39.5% |
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| GDP.pdf | — | 19.8% |
| METR Time Horizons | 62% | — |
Reasoning DeepSeek V4.1 Flash leads
Claude Sonnet 4: 22.9 (#187), DeepSeek V4.1 Flash: 50.2 (#36)
| Benchmark | Claude Sonnet 4 | DeepSeek V4.1 Flash |
|---|---|---|
| CritPt | 0.3% | 14.3% |
| LMArena Hard Prompts | 1372 | 1483 |
| DTBench | 77.1% | 89.9% |
| LMCA | 29% | 47% |
| Epoch Capabilities Index | 141.69 | 154.9 |
| ARC-AGI-2 | 5.9% | — |
| SimpleBench | 45.5% | — |
| Kagi LLM Benchmark | 73% | — |
| NYT Connections (extended) | — | 89.6% |
| ARC-AGI-1 | 40% | — |
| EnigmaEval | 3.1% | — |
| Mystery Game Puzzles | — | 43% |
| Surface Evolver Bench | — | 46.3% |
| ForecastBench | 60.2 | — |
Math DeepSeek V4.1 Flash leads
Claude Sonnet 4: 43.3 (#80), DeepSeek V4.1 Flash: 66.7 (#25)
| Benchmark | Claude Sonnet 4 | DeepSeek V4.1 Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 98.3% |
| LMArena Math | 1375 | 1477 |
| FrontierMath (Tiers 1-3) | — | 67.4% |
| FrontierMath Tier 4 | — | 26.8% |
| ProofBench | — | 54% |
| Omni-MATH | 60.2% | — |
| MATH Level 5 | 84.4% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge DeepSeek V4.1 Flash leads
Claude Sonnet 4: 41.8 (#108), DeepSeek V4.1 Flash: 57.9 (#38)
| Benchmark | Claude Sonnet 4 | DeepSeek V4.1 Flash |
|---|---|---|
| GPQA Diamond | 79.2% | 89.8% |
| LMArena Expert | 1372 | 1506 |
| Humanity's Last Exam | 7.8% | — |
| MMLU-Pro | 84.3% | — |
| Confabulations | 13.2% | — |
| Vectara Hallucination Rate | 10.3% | — |
| GPQA (HELM) | 70.6% | — |
Multimodal DeepSeek V4.1 Flash leads
Claude Sonnet 4: 26.2 (#121), DeepSeek V4.1 Flash: 39.1 (#61)
| Benchmark | Claude Sonnet 4 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Vision | 1191 | 1277 |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| Furniture Assembly | — | 34.2% |
| MindCube | 44.8% | — |
Multilingual DeepSeek V4.1 Flash leads
Claude Sonnet 4: 46.7 (#156), DeepSeek V4.1 Flash: 55.0 (#35)
| Benchmark | Claude Sonnet 4 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Non-English | 1333 | 1448 |
| LMArena Chinese | 1350 | 1497 |
| LMArena French | 1363 | 1452 |
| LMArena German | 1331 | 1484 |
| LMArena Japanese | 1302 | 1412 |
| LMArena Korean | 1291 | 1452 |
| LMArena Russian | 1355 | 1471 |
| LMArena Spanish | 1357 | 1459 |
Instruction Following DeepSeek V4.1 Flash leads
Claude Sonnet 4: 71.7 (#145), DeepSeek V4.1 Flash: 77.3 (#26)
| Benchmark | Claude Sonnet 4 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Instruction Following | 1376 | 1474 |
| IFEval | 84% | — |
Long Context DeepSeek V4.1 Flash leads
Claude Sonnet 4: 33.7 (#259), DeepSeek V4.1 Flash: 45.2 (#47)
| Benchmark | Claude Sonnet 4 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Longer Query | 1398 | 1475 |
| Fiction.LiveBench | 46.9% | — |
Writing & Preference DeepSeek V4.1 Flash leads
Claude Sonnet 4: 57.1 (#132), DeepSeek V4.1 Flash: 65.4 (#48)
| Benchmark | Claude Sonnet 4 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Text | 1351 | 1462 |
| LMArena Creative Writing | 1345 | 1435 |
| EQ-Bench Creative Writing | 1483 | 1540 |
| LMArena Multi-Turn | 1376 | 1457 |
| Short-Story Creative Writing | 81.4% | — |
| WildBench | 83.8% | — |
Frequently asked questions
Is Claude Sonnet 4 better than DeepSeek V4.1 Flash?
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 40.8 on the Noometry Index.
Which is cheaper, Claude Sonnet 4 or DeepSeek V4.1 Flash?
DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Claude Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or DeepSeek V4.1 Flash better for coding?
DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 43.5 in the Noometry coding category.
Which has the bigger context window?
DeepSeek V4.1 Flash does, with 1M tokens against 200K.
How many benchmarks do Claude Sonnet 4 and DeepSeek V4.1 Flash share?
27 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and DeepSeek V4.1 Flash has 37.