Model comparison
Claude Sonnet 5 vs DeepSeek V4 Pro
Claude Sonnet 5 and DeepSeek V4 Pro score almost the same on the Noometry Index (54.6 vs 54.3), so choose on price, context window or the category you care about most.
Last verified . 41 shared benchmarks.
DeepSeekSummary
- They share 41 benchmarks with published results for both. Claude Sonnet 5 scores higher in 5 categories and DeepSeek V4 Pro in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Claude Sonnet 5 leads 42.8 to 32.8.
- The biggest single-benchmark swing is ProofBench: 77% for Claude Sonnet 5 and 50% for DeepSeek V4 Pro.
- DeepSeek V4 Pro is cheaper at $0.66 / $1.98 per million input/output tokens, against $2 / $10 for Claude Sonnet 5.
- DeepSeek V4 Pro has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 5 | DeepSeek V4 Pro | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 54.6 | 54.3 |
| Released | 2026-06-29 | 2026-04-24 |
| Weights | Proprietary | Open |
| Context window | 1M | 1M |
| Max output | 128K | 393K |
| Input $ / M tokens | $2 | $0.66 |
| Output $ / M tokens | $10 | $1.98 |
| Results tracked | 51 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 5 leads
Claude Sonnet 5: 55.5 (#26), DeepSeek V4 Pro: 52.4 (#34)
| Benchmark | Claude Sonnet 5 | DeepSeek V4 Pro |
|---|---|---|
| FrontierCode | 42.7% | 28.6% |
| LMArena WebDev | 1541 | 1582 |
| SciCode | 54.3% | 51% |
| WeirdML | 68.8% | 66.2% |
| LMArena Coding | 1483 | 1470 |
| ALE-Bench | 1,463 | 1,403 |
| SWE-bench Verified | — | 77.6% |
| DeepSWE | 53.8% | — |
| CursorBench | 34.1% | — |
| GSO | 37.3% | — |
Agentic & Tool Use Claude Sonnet 5 leads
Claude Sonnet 5: 42.8 (#18), DeepSeek V4 Pro: 32.8 (#58)
| Benchmark | Claude Sonnet 5 | DeepSeek V4 Pro |
|---|---|---|
| APEX-Agents | 54.5% | 47.3% |
| Vending-Bench 2 | 6,378 | 3,285 |
| GBAEval | 65.3% | — |
| LMArena Search | 1194 | — |
Reasoning DeepSeek V4 Pro leads
Claude Sonnet 5: 49.1 (#39), DeepSeek V4 Pro: 56.5 (#24)
| Benchmark | Claude Sonnet 5 | DeepSeek V4 Pro |
|---|---|---|
| NYT Connections (extended) | 75.1% | 91.3% |
| CritPt | 16.9% | 18% |
| Chess Puzzles | 35% | 47% |
| LMArena Hard Prompts | 1461 | 1461 |
| Mystery Game Puzzles | 35% | 43% |
| DTBench | 92.5% | 93.9% |
| LMCA | 50% | 45.5% |
| Surface Evolver Bench | 60% | 40% |
| Epoch Capabilities Index | 156.21 | 155.31 |
| ForecastBench | 61.1 | 56.1 |
| ARC-AGI-2 | — | 61.3% |
| SimpleBench | 60.6% | — |
| Kagi LLM Benchmark | — | 53.5% |
| ARC-AGI-1 | — | 90.5% |
| Bench to the Future 3 | 0.14 | — |
Math Claude Sonnet 5 leads
Claude Sonnet 5: 66.2 (#27), DeepSeek V4 Pro: 64.8 (#30)
| Benchmark | Claude Sonnet 5 | DeepSeek V4 Pro |
|---|---|---|
| FrontierMath (Tiers 1-3) | 65.6% | 64.6% |
| FrontierMath Tier 4 | 29.3% | 26.8% |
| OTIS Mock AIME 2024-2025 | 94.7% | 98.6% |
| ProofBench | 77% | 50% |
| LMArena Math | 1467 | 1455 |
| MathArena Final-Answer Competitions | — | 76.6% |
Knowledge DeepSeek V4 Pro leads
Claude Sonnet 5: 55.6 (#47), DeepSeek V4 Pro: 59.5 (#31)
| Benchmark | Claude Sonnet 5 | DeepSeek V4 Pro |
|---|---|---|
| GPQA Diamond | 90.5% | 91.7% |
| SimpleQA Verified | 33.7% | 52.9% |
| LMArena Expert | 1490 | 1464 |
| Vectara Hallucination Rate | — | 8.6% |
Multimodal Not comparable
Claude Sonnet 5: 42.4 (#31), DeepSeek V4 Pro: —
| Benchmark | Claude Sonnet 5 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Vision | 1274 | — |
| Blueprint-Bench 2 | 24.9% | — |
| LMArena Document | 1466 | — |
Multilingual Too close to call
Claude Sonnet 5: 53.8 (#55), DeepSeek V4 Pro: 54.4 (#45)
| Benchmark | Claude Sonnet 5 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Non-English | 1431 | 1439 |
| LMArena Chinese | 1477 | 1486 |
| LMArena French | 1460 | 1472 |
| LMArena German | 1440 | 1458 |
| LMArena Japanese | 1422 | 1445 |
| LMArena Korean | 1411 | 1447 |
| LMArena Russian | 1451 | 1453 |
| LMArena Spanish | 1437 | 1458 |
Instruction Following Too close to call
Claude Sonnet 5: 76.3 (#41), DeepSeek V4 Pro: 76.1 (#47)
| Benchmark | Claude Sonnet 5 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Instruction Following | 1452 | 1448 |
Long Context Too close to call
Claude Sonnet 5: 44.8 (#55), DeepSeek V4 Pro: 45.0 (#51)
| Benchmark | Claude Sonnet 5 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Longer Query | 1463 | 1458 |
| CL-bench Life | — | 13.5% |
Writing & Preference Claude Sonnet 5 leads
Claude Sonnet 5: 69.2 (#25), DeepSeek V4 Pro: 65.5 (#46)
| Benchmark | Claude Sonnet 5 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Text | 1442 | 1451 |
| LMArena Creative Writing | 1416 | 1446 |
| EQ-Bench Creative Writing | 1794 | 1553 |
| EQ-Bench 4 | 1236 | 1166 |
| LMArena Multi-Turn | 1454 | 1467 |
Frequently asked questions
Is Claude Sonnet 5 better than DeepSeek V4 Pro?
Claude Sonnet 5 and DeepSeek V4 Pro score almost the same on the Noometry Index (54.6 vs 54.3), so choose on price, context window or the category you care about most.
Which is cheaper, Claude Sonnet 5 or DeepSeek V4 Pro?
DeepSeek V4 Pro is cheaper. It lists at $0.66 per million input tokens and $1.98 per million output tokens; Claude Sonnet 5 lists at $2 and $10.
Is Claude Sonnet 5 or DeepSeek V4 Pro better for coding?
Claude Sonnet 5 scores higher on coding benchmarks: 55.5 versus 52.4 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do Claude Sonnet 5 and DeepSeek V4 Pro share?
41 benchmarks have published results for both models. Claude Sonnet 5 has 51 scored results on Noometry and DeepSeek V4 Pro has 48.