Model comparison
Claude Sonnet 5 vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 54.6 on the Noometry Index.
Last verified . 35 shared benchmarks.
Alibaba (Qwen)Summary
- They share 35 benchmarks with published results for both. Claude Sonnet 5 scores higher in 3 categories and Qwen3.8 Max in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.2 to 66.2.
- The biggest single-benchmark swing is ProofBench: 77% for Claude Sonnet 5 and 58% for Qwen3.8 Max.
- Qwen3.8 Max is cheaper at $2 / $6 per million input/output tokens, against $2 / $10 for Claude Sonnet 5.
Side by side
| Claude Sonnet 5 | Qwen3.8 Max | |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 54.6 | 56.8 |
| Released | 2026-06-29 | 2026-08-02 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1M |
| Max output | 128K | 131K |
| Input $ / M tokens | $2 | $2 |
| Output $ / M tokens | $10 | $6 |
| Results tracked | 51 | 39 |
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), Qwen3.8 Max: 53.5 (#29)
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max |
|---|---|---|
| DeepSWE | 53.8% | 57.5% |
| LMArena WebDev | 1541 | 1674 |
| SciCode | 54.3% | 53.2% |
| LMArena Coding | 1483 | 1502 |
| FrontierCode | 42.7% | — |
| CursorBench | 34.1% | — |
| FrontierSWE | — | 17.8% |
| GSO | 37.3% | — |
| WeirdML | 68.8% | — |
| ALE-Bench | 1,463 | — |
Agentic & Tool Use Qwen3.8 Max leads
Claude Sonnet 5: 42.8 (#18), Qwen3.8 Max: 45.4 (#14)
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | 54.5% | 63.3% |
| τ²-bench Banking | — | 55.1% |
| GBAEval | 65.3% | — |
| GDP.pdf | — | 23.2% |
| LMArena Search | 1194 | — |
| Vending-Bench 2 | 6,378 | — |
Reasoning Qwen3.8 Max leads
Claude Sonnet 5: 49.1 (#39), Qwen3.8 Max: 54.4 (#26)
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max |
|---|---|---|
| NYT Connections (extended) | 75.1% | 88.3% |
| CritPt | 16.9% | 20% |
| Chess Puzzles | 35% | 40% |
| LMArena Hard Prompts | 1461 | 1496 |
| Mystery Game Puzzles | 35% | 38% |
| DTBench | 92.5% | 92% |
| LMCA | 50% | 46.2% |
| Epoch Capabilities Index | 156.21 | 156.41 |
| SimpleBench | 60.6% | — |
| Surface Evolver Bench | 60% | — |
| Bench to the Future 3 | 0.14 | — |
| ForecastBench | 61.1 | — |
Math Qwen3.8 Max leads
Claude Sonnet 5: 66.2 (#27), Qwen3.8 Max: 73.2 (#20)
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 65.6% | 74.7% |
| FrontierMath Tier 4 | 29.3% | 46.3% |
| OTIS Mock AIME 2024-2025 | 94.7% | 100% |
| ProofBench | 77% | 58% |
| LMArena Math | 1467 | 1499 |
Knowledge Qwen3.8 Max leads
Claude Sonnet 5: 55.6 (#47), Qwen3.8 Max: 61.7 (#27)
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 90.5% | 92.7% |
| SimpleQA Verified | 33.7% | 47.3% |
| LMArena Expert | 1490 | 1507 |
Multimodal Claude Sonnet 5 leads
Claude Sonnet 5: 42.4 (#31), Qwen3.8 Max: 37.2 (#75)
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | 1274 | 1314 |
| Blueprint-Bench 2 | 24.9% | — |
| Furniture Assembly | — | 20% |
| LMArena Document | 1466 | — |
Multilingual Qwen3.8 Max leads
Claude Sonnet 5: 53.8 (#55), Qwen3.8 Max: 56.7 (#18)
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1431 | 1472 |
| LMArena Chinese | 1477 | 1538 |
| LMArena French | 1460 | 1503 |
| LMArena German | 1440 | 1483 |
| LMArena Japanese | 1422 | 1467 |
| LMArena Korean | 1411 | 1461 |
| LMArena Russian | 1451 | 1481 |
| LMArena Spanish | 1437 | 1492 |
Instruction Following Qwen3.8 Max leads
Claude Sonnet 5: 76.3 (#41), Qwen3.8 Max: 77.6 (#17)
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1452 | 1479 |
Long Context Too close to call
Claude Sonnet 5: 44.8 (#55), Qwen3.8 Max: 45.6 (#31)
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1463 | 1489 |
Writing & Preference Claude Sonnet 5 leads
Claude Sonnet 5: 69.2 (#25), Qwen3.8 Max: 67.1 (#30)
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1442 | 1483 |
| LMArena Creative Writing | 1416 | 1479 |
| LMArena Multi-Turn | 1454 | 1489 |
| EQ-Bench Creative Writing | 1794 | — |
| EQ-Bench 4 | 1236 | — |
Frequently asked questions
Is Claude Sonnet 5 better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 54.6 on the Noometry Index.
Which is cheaper, Claude Sonnet 5 or Qwen3.8 Max?
Qwen3.8 Max is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; Claude Sonnet 5 lists at $2 and $10.
Is Claude Sonnet 5 or Qwen3.8 Max better for coding?
Claude Sonnet 5 scores higher on coding benchmarks: 55.5 versus 53.5 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do Claude Sonnet 5 and Qwen3.8 Max share?
35 benchmarks have published results for both models. Claude Sonnet 5 has 51 scored results on Noometry and Qwen3.8 Max has 39.