Model comparison
Claude Sonnet 4.6 vs DeepSeek V4.1 Flash
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 50.3 on the Noometry Index.
Last verified . 33 shared benchmarks.
DeepSeekSummary
- They share 33 benchmarks with published results for both. Claude Sonnet 4.6 scores higher in 4 categories and DeepSeek V4.1 Flash in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4.1 Flash leads 66.7 to 52.9.
- The biggest single-benchmark swing is Mystery Game Puzzles: 16% for Claude Sonnet 4.6 and 43% 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 4.6.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4.6 | DeepSeek V4.1 Flash | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 50.3 | 52.8 |
| Released | 2026-02-17 | 2026-09-09 |
| Weights | Proprietary | Open |
| Context window | 1M | 1M |
| Max output | 128K | 393K |
| Input $ / M tokens | $3 | $0.15 |
| Output $ / M tokens | $15 | $0.60 |
| Results tracked | 57 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4.1 Flash leads
Claude Sonnet 4.6: 46.3 (#67), DeepSeek V4.1 Flash: 52.9 (#32)
| Benchmark | Claude Sonnet 4.6 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena WebDev | 1522 | 1619 |
| SciCode | 46.8% | 51.9% |
| LMArena Coding | 1504 | 1506 |
| ALE-Bench | 1,327 | 1,092 |
| SWE-bench Verified | 75.2% | — |
| DeepSWE | 29.9% | — |
| FrontierCode | 24.3% | — |
| WeirdML | 66.1% | — |
Agentic & Tool Use Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 39.1 (#28), DeepSeek V4.1 Flash: 31.2 (#69)
| Benchmark | Claude Sonnet 4.6 | DeepSeek V4.1 Flash |
|---|---|---|
| APEX-Agents | 43% | 39.5% |
| GDP.pdf | 18% | 19.8% |
| Terminal-Bench | 53.4% | — |
| OSWorld 2.0 | 9.3% | — |
| DeepResearch Bench | 54.9% | — |
| OSWorld | 72.1% | — |
| ExploitBench | 23.6% | — |
| GBAEval | 48.8% | — |
| LMArena Search | 1221 | — |
| Vending-Bench 2 | 7,204 | — |
Reasoning DeepSeek V4.1 Flash leads
Claude Sonnet 4.6: 46.1 (#45), DeepSeek V4.1 Flash: 50.2 (#36)
| Benchmark | Claude Sonnet 4.6 | DeepSeek V4.1 Flash |
|---|---|---|
| NYT Connections (extended) | 80.9% | 89.6% |
| CritPt | 3.1% | 14.3% |
| LMArena Hard Prompts | 1484 | 1483 |
| Mystery Game Puzzles | 16% | 43% |
| DTBench | 89.9% | 89.9% |
| LMCA | 46.5% | 47% |
| Epoch Capabilities Index | 152.24 | 154.9 |
| ARC-AGI-2 | 60.4% | — |
| ARC-AGI-1 | 86.5% | — |
| Chess Puzzles | 13% | — |
| Thematic Generalization | 76.3% | — |
| Surface Evolver Bench | — | 46.3% |
| ForecastBench | 62 | — |
Math DeepSeek V4.1 Flash leads
Claude Sonnet 4.6: 52.9 (#49), DeepSeek V4.1 Flash: 66.7 (#25)
| Benchmark | Claude Sonnet 4.6 | DeepSeek V4.1 Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 85.8% | 98.3% |
| ProofBench | 45% | 54% |
| LMArena Math | 1462 | 1477 |
| FrontierMath (Tiers 1-3) | — | 67.4% |
| FrontierMath Tier 4 | — | 26.8% |
| FrontierMath (Feb 2025 set) | 32.4% | — |
| FrontierMath Tier 4 (v1) | 8.3% | — |
Knowledge DeepSeek V4.1 Flash leads
Claude Sonnet 4.6: 51.7 (#65), DeepSeek V4.1 Flash: 57.9 (#38)
| Benchmark | Claude Sonnet 4.6 | DeepSeek V4.1 Flash |
|---|---|---|
| GPQA Diamond | 87.4% | 89.8% |
| LMArena Expert | 1500 | 1506 |
| SimpleQA Verified | 35.5% | — |
| Vectara Hallucination Rate | 10.6% | — |
Multimodal DeepSeek V4.1 Flash leads
Claude Sonnet 4.6: 38.0 (#68), DeepSeek V4.1 Flash: 39.1 (#61)
| Benchmark | Claude Sonnet 4.6 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Vision | 1283 | 1277 |
| Blueprint-Bench 2 | 6.7% | — |
| Furniture Assembly | — | 34.2% |
| LMArena Document | 1482 | — |
Multilingual Too close to call
Claude Sonnet 4.6: 54.4 (#41), DeepSeek V4.1 Flash: 55.0 (#35)
| Benchmark | Claude Sonnet 4.6 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Non-English | 1440 | 1448 |
| LMArena Chinese | 1491 | 1497 |
| LMArena French | 1465 | 1452 |
| LMArena German | 1428 | 1484 |
| LMArena Japanese | 1420 | 1412 |
| LMArena Korean | 1411 | 1452 |
| LMArena Russian | 1440 | 1471 |
| LMArena Spanish | 1464 | 1459 |
Instruction Following Too close to call
Claude Sonnet 4.6: 77.4 (#25), DeepSeek V4.1 Flash: 77.3 (#26)
| Benchmark | Claude Sonnet 4.6 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Instruction Following | 1475 | 1474 |
Long Context Too close to call
Claude Sonnet 4.6: 45.3 (#44), DeepSeek V4.1 Flash: 45.2 (#47)
| Benchmark | Claude Sonnet 4.6 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Longer Query | 1479 | 1475 |
Writing & Preference Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 70.2 (#22), DeepSeek V4.1 Flash: 65.4 (#48)
| Benchmark | Claude Sonnet 4.6 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Text | 1458 | 1462 |
| LMArena Creative Writing | 1435 | 1435 |
| EQ-Bench Creative Writing | 1810 | 1540 |
| LMArena Multi-Turn | 1464 | 1457 |
| EQ-Bench 4 | 1207 | — |
Frequently asked questions
Is Claude Sonnet 4.6 better than DeepSeek V4.1 Flash?
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 50.3 on the Noometry Index.
Which is cheaper, Claude Sonnet 4.6 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 4.6 lists at $3 and $15.
Is Claude Sonnet 4.6 or DeepSeek V4.1 Flash better for coding?
DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 46.3 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do Claude Sonnet 4.6 and DeepSeek V4.1 Flash share?
33 benchmarks have published results for both models. Claude Sonnet 4.6 has 57 scored results on Noometry and DeepSeek V4.1 Flash has 37.