Model comparison
Claude Opus 4.5 vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 50.5 on the Noometry Index.
Last verified . 32 shared benchmarks.
Alibaba (Qwen)Summary
- They share 32 benchmarks with published results for both. Claude Opus 4.5 scores higher in 4 categories and Qwen3.8 Max in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.2 to 38.6.
- The biggest single-benchmark swing is FrontierMath Tier 4: 4.9% for Claude Opus 4.5 and 46.3% for Qwen3.8 Max.
- Qwen3.8 Max is cheaper at $2 / $6 per million input/output tokens, against $5 / $25 for Claude Opus 4.5.
- Qwen3.8 Max accepts more context: 1M tokens versus 200K.
Side by side
| Claude Opus 4.5 | Qwen3.8 Max | |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 50.5 | 56.8 |
| Released | 2025-11-01 | 2026-08-02 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 1M |
| Max output | 64K | 131K |
| Input $ / M tokens | $5 | $2 |
| Output $ / M tokens | $25 | $6 |
| Results tracked | 69 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4.5 leads
Claude Opus 4.5: 54.8 (#27), Qwen3.8 Max: 53.5 (#29)
| Benchmark | Claude Opus 4.5 | Qwen3.8 Max |
|---|---|---|
| LMArena WebDev | 1494 | 1674 |
| LMArena Coding | 1504 | 1502 |
| SWE-bench Verified | 76.7% | — |
| DeepSWE | — | 57.5% |
| SWE-bench Verified (bash only) | 76.8% | — |
| SWE-bench Multilingual | 70.7% | — |
| FrontierSWE | — | 17.8% |
| SciCode | — | 53.2% |
| GSO | 26.5% | — |
| WeirdML | 63.7% | — |
| ALE-Bench | 1,025 | — |
| AlgoTune | 1.77 | — |
Agentic & Tool Use Claude Opus 4.5 leads
Claude Opus 4.5: 47.3 (#12), Qwen3.8 Max: 45.4 (#14)
| Benchmark | Claude Opus 4.5 | Qwen3.8 Max |
|---|---|---|
| τ²-bench Banking | 24.7% | 55.1% |
| Terminal-Bench | 63.1% | — |
| APEX-Agents | — | 63.3% |
| Berkeley Function Calling Leaderboard | 77.5% | — |
| GDPval | 45.5% | — |
| Remote Labor Index | 3.8% | — |
| τ²-bench Airline | 84% | — |
| τ²-bench Retail | 79.6% | — |
| τ²-bench Telecom | 92.3% | — |
| Cybench | 82% | — |
| DeepResearch Bench | 54.8% | — |
| OSWorld | 66.3% | — |
| BALROG | 43.5% | — |
| GDP.pdf | — | 23.2% |
| LMArena Search | 1180 | — |
| METR Time Horizons | 75% | — |
| Vending-Bench 2 | 4,967 | — |
Reasoning Qwen3.8 Max leads
Claude Opus 4.5: 42.6 (#51), Qwen3.8 Max: 54.4 (#26)
| Benchmark | Claude Opus 4.5 | Qwen3.8 Max |
|---|---|---|
| NYT Connections (extended) | 52.5% | 88.3% |
| Chess Puzzles | 12% | 40% |
| LMArena Hard Prompts | 1476 | 1496 |
| Mystery Game Puzzles | 22% | 38% |
| DTBench | 89.9% | 92% |
| LMCA | 44.5% | 46.2% |
| Epoch Capabilities Index | 150.09 | 156.41 |
| ARC-AGI-2 | 37.6% | — |
| SimpleBench | 62% | — |
| Kagi LLM Benchmark | 80.2% | — |
| ARC-AGI-1 | 80% | — |
| CritPt | — | 20% |
| EnigmaEval | 11.9% | — |
| EBR-Bench | 14.3% | — |
| ForecastBench | 60.7 | — |
Math Qwen3.8 Max leads
Claude Opus 4.5: 38.6 (#132), Qwen3.8 Max: 73.2 (#20)
| Benchmark | Claude Opus 4.5 | Qwen3.8 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 34.4% | 74.7% |
| FrontierMath Tier 4 | 4.9% | 46.3% |
| OTIS Mock AIME 2024-2025 | 86.1% | 100% |
| ProofBench | 36% | 58% |
| LMArena Math | 1463 | 1499 |
| FrontierMath (Feb 2025 set) | 20.7% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Qwen3.8 Max leads
Claude Opus 4.5: 56.5 (#44), Qwen3.8 Max: 61.7 (#27)
| Benchmark | Claude Opus 4.5 | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 86% | 92.7% |
| SimpleQA Verified | 45.7% | 47.3% |
| LMArena Expert | 1487 | 1507 |
| Humanity's Last Exam | 25.2% | — |
| Vectara Hallucination Rate | 10.9% | — |
Multimodal Qwen3.8 Max leads
Claude Opus 4.5: 31.4 (#107), Qwen3.8 Max: 37.2 (#75)
| Benchmark | Claude Opus 4.5 | Qwen3.8 Max |
|---|---|---|
| Furniture Assembly | 28.3% | 20% |
| LMArena Vision | — | 1314 |
| GeoBench | 75% | — |
| VPCT | 40% | — |
| LMArena Document | 1462 | — |
Multilingual Qwen3.8 Max leads
Claude Opus 4.5: 54.3 (#47), Qwen3.8 Max: 56.7 (#18)
| Benchmark | Claude Opus 4.5 | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1438 | 1472 |
| LMArena Chinese | 1470 | 1538 |
| LMArena French | 1471 | 1503 |
| LMArena German | 1449 | 1483 |
| LMArena Japanese | 1416 | 1467 |
| LMArena Korean | 1424 | 1461 |
| LMArena Russian | 1447 | 1481 |
| LMArena Spanish | 1458 | 1492 |
Instruction Following Too close to call
Claude Opus 4.5: 77.5 (#19), Qwen3.8 Max: 77.6 (#17)
| Benchmark | Claude Opus 4.5 | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1478 | 1479 |
Long Context Too close to call
Claude Opus 4.5: 46.5 (#22), Qwen3.8 Max: 45.6 (#31)
| Benchmark | Claude Opus 4.5 | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1480 | 1489 |
| CL-bench | 21.1% | — |
Writing & Preference Too close to call
Claude Opus 4.5: 68.1 (#28), Qwen3.8 Max: 67.1 (#30)
| Benchmark | Claude Opus 4.5 | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1451 | 1483 |
| LMArena Creative Writing | 1445 | 1479 |
| LMArena Multi-Turn | 1466 | 1489 |
| EQ-Bench Creative Writing | 1687 | — |
Frequently asked questions
Is Claude Opus 4.5 better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 50.5 on the Noometry Index.
Which is cheaper, Claude Opus 4.5 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 4.5 lists at $5 and $25.
Is Claude Opus 4.5 or Qwen3.8 Max better for coding?
Claude Opus 4.5 scores higher on coding benchmarks: 54.8 versus 53.5 in the Noometry coding category.
Which has the bigger context window?
Qwen3.8 Max does, with 1M tokens against 200K.
How many benchmarks do Claude Opus 4.5 and Qwen3.8 Max share?
32 benchmarks have published results for both models. Claude Opus 4.5 has 69 scored results on Noometry and Qwen3.8 Max has 39.