Leaderboard / Model
Qwen 3.8 Max
Alibaba · Closed weights · Released Aug 2, 2026 · Price not listed
n/aOverall · not enough data
Score by job
Coding
Not enough data · 2 of 5 benchmarks
| Benchmark | Result | Variant | z |
|---|---|---|---|
| DeepSWE | 57.5% | qwen3.8-max_xhigh | 0.02 |
| LMArena Coding | 1500 rating | qwen3.8-max | 0.46 |
Agents
Not enough data · 0 of 4 benchmarks
Reasoning
Not enough data · 1 of 4 benchmarks
| Benchmark | Result | Variant | z |
|---|---|---|---|
| GPQA Diamond | 92.7% | qwen3.8-max_xhigh | 0.45 |
Math
62 · rank 10 · 5 of 5 benchmarks
| Benchmark | Result | Variant | z |
|---|---|---|---|
| FrontierMath Tiers 1–3 | 74.7% | qwen3.8-max_xhigh | 0.47 |
| FrontierMath Tier 4 | 46.3% | qwen3.8-max_xhigh | 0.22 |
| OTIS Mock AIME | 99.4% | qwen3.8-max_xhigh | 0.95 |
| ProofBench | 58.0% | qwen3.8-max_unknown | 0.41 |
| LMArena Math | 1500 rating | qwen3.8-max | 0.76 |
Writing
78 · rank 8 · 2 of 2 benchmarks
| Benchmark | Result | Variant | z |
|---|---|---|---|
| LMArena Creative Writing | 1481 rating | qwen3.8-max | 0.91 |
| LMArena Instruction Following | 1482 rating | qwen3.8-max | 0.67 |
Design
67 · rank 6 · 1 of 1 benchmarks
| Benchmark | Result | Variant | z |
|---|---|---|---|
| LMArena WebDev | 1671 rating | qwen3.8-max | 1.12 |