Model comparison
Claude Sonnet 4.6 vs Deepseek Coder v2
Claude Sonnet 4.6 is the stronger model overall, scoring 50.1 to 35.3 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeekSummary
- They share 17 benchmarks with published results for both. Claude Sonnet 4.6 scores higher in 8 categories and Deepseek Coder v2 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Claude Sonnet 4.6 leads 70.4 to 38.4.
- Deepseek Coder v2 has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4.6 | Deepseek Coder v2 | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 50.1 | 35.3 |
| Released | 2026-02-17 | 2024-06-17 |
| Weights | Proprietary | Open |
| Context window | 1M | — |
| Max output | 128K | — |
| Input $ / M tokens | $3 | — |
| Output $ / M tokens | $15 | — |
| Results tracked | 57 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 46.0 (#62), Deepseek Coder v2: 37.5 (#178)
| Benchmark | Claude Sonnet 4.6 | Deepseek Coder v2 |
|---|---|---|
| LMArena Coding | 1504 | 1251 |
| SWE-bench Verified | 75.2% | — |
| DeepSWE | 29.9% | — |
| FrontierCode | 24.3% | — |
| LMArena WebDev | 1521 | — |
| SciCode | 46.8% | — |
| WeirdML | 66.1% | — |
| BigCodeBench Instruct | — | 48.2% |
| BigCodeBench Complete | — | 59.7% |
| ALE-Bench | 1,327 | — |
| HumanEval+ | — | 82.3% |
| MBPP+ | — | 75.1% |
Agentic & Tool Use Not comparable
Claude Sonnet 4.6: 39.1 (#26), Deepseek Coder v2: —
| Benchmark | Claude Sonnet 4.6 | Deepseek Coder v2 |
|---|---|---|
| Terminal-Bench | 53.4% | — |
| APEX-Agents | 43% | — |
| OSWorld 2.0 | 9.3% | — |
| DeepResearch Bench | 54.9% | — |
| OSWorld | 72.1% | — |
| ExploitBench | 23.6% | — |
| GBAEval | 48.8% | — |
| GDP.pdf | 18% | — |
| LMArena Search | 1221 | — |
| Vending-Bench 2 | 7,204 | — |
Reasoning Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 46.1 (#40), Deepseek Coder v2: 22.6 (#175)
| Benchmark | Claude Sonnet 4.6 | Deepseek Coder v2 |
|---|---|---|
| LMArena Hard Prompts | 1485 | 1207 |
| ARC-AGI-2 | 60.4% | — |
| NYT Connections (extended) | 80.9% | — |
| ARC-AGI-1 | 86.5% | — |
| CritPt | 3.1% | — |
| Chess Puzzles | 13% | — |
| Thematic Generalization | 76.3% | — |
| Mystery Game Puzzles | 16% | — |
| DTBench | 89.9% | — |
| LMCA | 46.5% | — |
| Epoch Capabilities Index | 152.25 | — |
| ForecastBench | 62 | — |
| WinoGrande | — | 83.7% |
Math Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 51.9 (#41), Deepseek Coder v2: 32.3 (#188)
| Benchmark | Claude Sonnet 4.6 | Deepseek Coder v2 |
|---|---|---|
| LMArena Math | 1463 | 1241 |
| OTIS Mock AIME 2024-2025 | 85.8% | — |
| ProofBench | 45% | — |
| FrontierMath (Feb 2025 set) | 32.4% | — |
| FrontierMath Tier 4 (v1) | 8.3% | — |
| GSM8K | — | 94.5% |
Knowledge Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 51.6 (#60), Deepseek Coder v2: 31.7 (#209)
| Benchmark | Claude Sonnet 4.6 | Deepseek Coder v2 |
|---|---|---|
| LMArena Expert | 1500 | 1181 |
| GPQA Diamond | 87.4% | — |
| SimpleQA Verified | 35.5% | — |
| Vectara Hallucination Rate | 10.6% | — |
| ARC (AI2) Challenge | — | 64.3% |
Multimodal Not comparable
Claude Sonnet 4.6: 38.0 (#61), Deepseek Coder v2: —
| Benchmark | Claude Sonnet 4.6 | Deepseek Coder v2 |
|---|---|---|
| LMArena Vision | 1283 | — |
| Blueprint-Bench 2 | 6.7% | — |
| LMArena Document | 1482 | — |
Multilingual Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 54.6 (#39), Deepseek Coder v2: 36.4 (#235)
| Benchmark | Claude Sonnet 4.6 | Deepseek Coder v2 |
|---|---|---|
| LMArena Non-English | 1440 | 1182 |
| LMArena Chinese | 1491 | 1201 |
| LMArena French | 1465 | 1184 |
| LMArena German | 1428 | 1164 |
| LMArena Japanese | 1419 | 1126 |
| LMArena Korean | 1411 | 1104 |
| LMArena Russian | 1440 | 1188 |
| LMArena Spanish | 1465 | 1153 |
Instruction Following Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 77.6 (#23), Deepseek Coder v2: 61.9 (#237)
| Benchmark | Claude Sonnet 4.6 | Deepseek Coder v2 |
|---|---|---|
| LMArena Instruction Following | 1476 | 1180 |
Long Context Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 45.3 (#40), Deepseek Coder v2: 37.0 (#219)
| Benchmark | Claude Sonnet 4.6 | Deepseek Coder v2 |
|---|---|---|
| LMArena Longer Query | 1479 | 1219 |
Writing & Preference Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 70.4 (#22), Deepseek Coder v2: 38.4 (#249)
| Benchmark | Claude Sonnet 4.6 | Deepseek Coder v2 |
|---|---|---|
| LMArena Text | 1458 | 1191 |
| LMArena Creative Writing | 1435 | 1120 |
| LMArena Multi-Turn | 1464 | 1176 |
| EQ-Bench Creative Writing | 1810 | — |
| EQ-Bench 4 | 1207 | — |
Frequently asked questions
Is Claude Sonnet 4.6 better than Deepseek Coder v2?
Claude Sonnet 4.6 is the stronger model overall, scoring 50.1 to 35.3 on the Noometry Index.
Is Claude Sonnet 4.6 or Deepseek Coder v2 better for coding?
Claude Sonnet 4.6 scores higher on coding benchmarks: 46.0 versus 37.5 in the Noometry coding category.
How many benchmarks do Claude Sonnet 4.6 and Deepseek Coder v2 share?
17 benchmarks have published results for both models. Claude Sonnet 4.6 has 57 scored results on Noometry and Deepseek Coder v2 has 24.