Model comparison
Claude Opus 5 vs Qwen3.8 Max
Claude Opus 5 is the stronger model overall, scoring 67.8 to 56.8 on the Noometry Index. Qwen3.8 Max costs 3.3× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Last verified . 39 shared benchmarks.
Alibaba (Qwen)Summary
- They share 39 benchmarks with published results for both. Claude Opus 5 scores higher in 10 categories and Qwen3.8 Max in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 5 leads 77.2 to 54.4.
- The biggest single-benchmark swing is ProofBench: 99% for Claude Opus 5 and 58% for Qwen3.8 Max.
- Qwen3.8 Max is cheaper at $2 / $6 per million input/output tokens, against $5 / $25 for Claude Opus 5.
Side by side
| Claude Opus 5 | Qwen3.8 Max | |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 67.8 | 56.8 |
| Released | 2026-07-24 | 2026-08-02 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1M |
| Max output | 128K | 131K |
| Input $ / M tokens | $5 | $2 |
| Output $ / M tokens | $25 | $6 |
| Results tracked | 57 | 39 |
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), Qwen3.8 Max: 53.5 (#29)
| Benchmark | Claude Opus 5 | Qwen3.8 Max |
|---|---|---|
| DeepSWE | 73.6% | 57.5% |
| LMArena WebDev | 1691 | 1674 |
| FrontierSWE | 52% | 17.8% |
| SciCode | 56.4% | 53.2% |
| LMArena Coding | 1534 | 1502 |
| FrontierCode | 53.4% | — |
| CursorBench | 46.6% | — |
| WeirdML | 91.8% | — |
| ALE-Bench | 2,165 | — |
Agentic & Tool Use Claude Opus 5 leads
Claude Opus 5: 55.6 (#1), Qwen3.8 Max: 45.4 (#14)
| Benchmark | Claude Opus 5 | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | 65.8% | 63.3% |
| τ²-bench Banking | 48.7% | 55.1% |
| GDP.pdf | 24% | 23.2% |
| OSWorld 2.0 | 31.4% | — |
| PostTrainBench | 35% | — |
| BALROG | 63.4% | — |
| GBAEval | 79.6% | — |
| Vending-Bench 2 | 11,182 | — |
Reasoning Claude Opus 5 leads
Claude Opus 5: 77.2 (#4), Qwen3.8 Max: 54.4 (#26)
| Benchmark | Claude Opus 5 | Qwen3.8 Max |
|---|---|---|
| NYT Connections (extended) | 94.3% | 88.3% |
| CritPt | 29.1% | 20% |
| Chess Puzzles | 42% | 40% |
| LMArena Hard Prompts | 1526 | 1496 |
| Mystery Game Puzzles | 59% | 38% |
| DTBench | 97.9% | 92% |
| LMCA | 64.5% | 46.2% |
| Epoch Capabilities Index | 162.78 | 156.41 |
| ARC-AGI-2 | 90.4% | — |
| SimpleBench | 80.6% | — |
| ARC-AGI-1 | 97.5% | — |
| EBR-Bench | 45.7% | — |
| Bench to the Future 3 | 0.12 | — |
Math Claude Opus 5 leads
Claude Opus 5: 86.2 (#8), Qwen3.8 Max: 73.2 (#20)
| Benchmark | Claude Opus 5 | Qwen3.8 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 85.6% | 74.7% |
| FrontierMath Tier 4 | 73.2% | 46.3% |
| OTIS Mock AIME 2024-2025 | 98.9% | 100% |
| ProofBench | 99% | 58% |
| LMArena Math | 1531 | 1499 |
Knowledge Claude Opus 5 leads
Claude Opus 5: 66.8 (#9), Qwen3.8 Max: 61.7 (#27)
| Benchmark | Claude Opus 5 | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 93.9% | 92.7% |
| SimpleQA Verified | 59.9% | 47.3% |
| LMArena Expert | 1557 | 1507 |
Multimodal Claude Opus 5 leads
Claude Opus 5: 50.8 (#8), Qwen3.8 Max: 37.2 (#75)
| Benchmark | Claude Opus 5 | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | 1319 | 1314 |
| Furniture Assembly | 60.8% | 20% |
| Blueprint-Bench 2 | 30.4% | — |
| LMArena Document | 1516 | — |
Multilingual Claude Opus 5 leads
Claude Opus 5: 58.8 (#4), Qwen3.8 Max: 56.7 (#18)
| Benchmark | Claude Opus 5 | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1501 | 1472 |
| LMArena Chinese | 1574 | 1538 |
| LMArena French | 1519 | 1503 |
| LMArena German | 1524 | 1483 |
| LMArena Japanese | 1516 | 1467 |
| LMArena Korean | 1521 | 1461 |
| LMArena Russian | 1507 | 1481 |
| LMArena Spanish | 1519 | 1492 |
Instruction Following Claude Opus 5 leads
Claude Opus 5: 79.2 (#7), Qwen3.8 Max: 77.6 (#17)
| Benchmark | Claude Opus 5 | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1517 | 1479 |
Long Context Too close to call
Claude Opus 5: 46.5 (#21), Qwen3.8 Max: 45.6 (#31)
| Benchmark | Claude Opus 5 | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1515 | 1489 |
Writing & Preference Claude Opus 5 leads
Claude Opus 5: 79.2 (#1), Qwen3.8 Max: 67.1 (#30)
| Benchmark | Claude Opus 5 | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1507 | 1483 |
| LMArena Creative Writing | 1491 | 1479 |
| LMArena Multi-Turn | 1499 | 1489 |
| EQ-Bench Creative Writing | 2133 | — |
| EQ-Bench 4 | 1385 | — |
Frequently asked questions
Is Claude Opus 5 better than Qwen3.8 Max?
Claude Opus 5 is the stronger model overall, scoring 67.8 to 56.8 on the Noometry Index. Qwen3.8 Max costs 3.3× 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 Qwen3.8 Max?
Qwen3.8 Max is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; Claude Opus 5 lists at $5 and $25.
Is Claude Opus 5 or Qwen3.8 Max better for coding?
Claude Opus 5 scores higher on coding benchmarks: 67.5 versus 53.5 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do Claude Opus 5 and Qwen3.8 Max share?
39 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and Qwen3.8 Max has 39.