Model comparison
Claude Opus 5 vs DeepSeek-V3.1
Claude Opus 5 is the stronger model overall, scoring 67.8 to 42.8 on the Noometry Index. DeepSeek-V3.1 costs 24× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Last verified . 23 shared benchmarks.
DeepSeekSummary
- They share 23 benchmarks with published results for both. Claude Opus 5 scores higher in 8 categories and DeepSeek-V3.1 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 5 leads 77.2 to 27.9.
- The biggest single-benchmark swing is WeirdML: 91.8% for Claude Opus 5 and 38.4% for DeepSeek-V3.1.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $5 / $25 for Claude Opus 5.
- Claude Opus 5 accepts more context: 1M tokens versus 164K.
- DeepSeek-V3.1 has downloadable open weights; the other is API-only.
Side by side
| Claude Opus 5 | DeepSeek-V3.1 | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 67.8 | 42.8 |
| Released | 2026-07-24 | 2025-08-21 |
| Weights | Proprietary | Open |
| Context window | 1M | 164K |
| Max output | 128K | 8K |
| Input $ / M tokens | $5 | $0.25 |
| Output $ / M tokens | $25 | $0.95 |
| Results tracked | 57 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 5 leads
Claude Opus 5: 67.5 (#5), DeepSeek-V3.1: 40.3 (#144)
| Benchmark | Claude Opus 5 | DeepSeek-V3.1 |
|---|---|---|
| WeirdML | 91.8% | 38.4% |
| LMArena Coding | 1534 | 1417 |
| DeepSWE | 73.6% | — |
| FrontierCode | 53.4% | — |
| CursorBench | 46.6% | — |
| LMArena WebDev | 1691 | — |
| FrontierSWE | 52% | — |
| SciCode | 56.4% | — |
| ALE-Bench | 2,165 | — |
Agentic & Tool Use Not comparable
Claude Opus 5: 55.6 (#1), DeepSeek-V3.1: —
| Benchmark | Claude Opus 5 | DeepSeek-V3.1 |
|---|---|---|
| APEX-Agents | 65.8% | — |
| OSWorld 2.0 | 31.4% | — |
| τ²-bench Banking | 48.7% | — |
| PostTrainBench | 35% | — |
| BALROG | 63.4% | — |
| GBAEval | 79.6% | — |
| GDP.pdf | 24% | — |
| Vending-Bench 2 | 11,182 | — |
Reasoning Claude Opus 5 leads
Claude Opus 5: 77.2 (#4), DeepSeek-V3.1: 27.9 (#110)
| Benchmark | Claude Opus 5 | DeepSeek-V3.1 |
|---|---|---|
| SimpleBench | 80.6% | 40% |
| LMArena Hard Prompts | 1526 | 1417 |
| DTBench | 97.9% | 82.7% |
| LMCA | 64.5% | 24.3% |
| Epoch Capabilities Index | 162.78 | 139.92 |
| ARC-AGI-2 | 90.4% | — |
| Kagi LLM Benchmark | — | 53.2% |
| NYT Connections (extended) | 94.3% | — |
| ARC-AGI-1 | 97.5% | — |
| CritPt | 29.1% | — |
| Chess Puzzles | 42% | — |
| EBR-Bench | 45.7% | — |
| Mystery Game Puzzles | 59% | — |
| Bench to the Future 3 | 0.12 | — |
| ForecastBench | — | 58 |
Math Claude Opus 5 leads
Claude Opus 5: 86.2 (#8), DeepSeek-V3.1: 38.9 (#122)
| Benchmark | Claude Opus 5 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Math | 1531 | 1420 |
| FrontierMath (Tiers 1-3) | 85.6% | — |
| FrontierMath Tier 4 | 73.2% | — |
| OTIS Mock AIME 2024-2025 | 98.9% | — |
| ProofBench | 99% | — |
Knowledge Claude Opus 5 leads
Claude Opus 5: 66.8 (#9), DeepSeek-V3.1: 43.7 (#90)
| Benchmark | Claude Opus 5 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Expert | 1557 | 1405 |
| GPQA Diamond | 93.9% | — |
| SimpleQA Verified | 59.9% | — |
| Vectara Hallucination Rate | — | 5.5% |
Multimodal Not comparable
Claude Opus 5: 50.8 (#8), DeepSeek-V3.1: —
| Benchmark | Claude Opus 5 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Vision | 1319 | — |
| Blueprint-Bench 2 | 30.4% | — |
| Furniture Assembly | 60.8% | — |
| LMArena Document | 1516 | — |
Multilingual Claude Opus 5 leads
Claude Opus 5: 58.8 (#4), DeepSeek-V3.1: 51.6 (#106)
| Benchmark | Claude Opus 5 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Non-English | 1501 | 1400 |
| LMArena Chinese | 1574 | 1469 |
| LMArena French | 1519 | 1447 |
| LMArena German | 1524 | 1411 |
| LMArena Japanese | 1516 | 1378 |
| LMArena Korean | 1521 | 1337 |
| LMArena Russian | 1507 | 1405 |
| LMArena Spanish | 1519 | 1431 |
Instruction Following Claude Opus 5 leads
Claude Opus 5: 79.2 (#7), DeepSeek-V3.1: 73.9 (#110)
| Benchmark | Claude Opus 5 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Instruction Following | 1517 | 1400 |
Long Context Claude Opus 5 leads
Claude Opus 5: 46.5 (#21), DeepSeek-V3.1: 36.3 (#232)
| Benchmark | Claude Opus 5 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Longer Query | 1515 | 1422 |
| Fiction.LiveBench | — | 52.8% |
Writing & Preference Claude Opus 5 leads
Claude Opus 5: 79.2 (#1), DeepSeek-V3.1: 60.3 (#98)
| Benchmark | Claude Opus 5 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Text | 1507 | 1420 |
| LMArena Creative Writing | 1491 | 1401 |
| EQ-Bench Creative Writing | 2133 | 1436 |
| LMArena Multi-Turn | 1499 | 1408 |
| EQ-Bench 4 | 1385 | — |
Frequently asked questions
Is Claude Opus 5 better than DeepSeek-V3.1?
Claude Opus 5 is the stronger model overall, scoring 67.8 to 42.8 on the Noometry Index. DeepSeek-V3.1 costs 24× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 5 or DeepSeek-V3.1?
DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; Claude Opus 5 lists at $5 and $25.
Is Claude Opus 5 or DeepSeek-V3.1 better for coding?
Claude Opus 5 scores higher on coding benchmarks: 67.5 versus 40.3 in the Noometry coding category.
Which has the bigger context window?
Claude Opus 5 does, with 1M tokens against 164K.
How many benchmarks do Claude Opus 5 and DeepSeek-V3.1 share?
23 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and DeepSeek-V3.1 has 27.