Qwen3
Qwen3.8 Max
newRank 8 of 45 · release 2026-06-25 · $2 in / $6 out per MTok
Global average
78.5
first appearance
Signal breakdown
Category scores
Reasoning
88.2
Coding
72.9
Agentic Coding
64.7
Mathematics
91.3
Data Analysis
78.4
Language
79.7
Instruction Following
74.1
Trend
Score history
Analyst annotations
Commentary
Sign in to publish an annotation under your analyst identity.
No annotations yet
Be the first analyst on record for Qwen3.8 Max.
Detail
Per-task scores
| Task | Category | Score |
|---|---|---|
| zebra_puzzle | 100.0 | |
| spatial | 100.0 | |
| tablereformat | 98.0 | |
| AMPS_Hard | 98.0 | |
| math_comp | 95.1 | |
| connections | 94.5 | |
| olympiad | 91.2 | |
| consecutive_events | 87.1 | |
| typos | 86.0 | |
| integrals_with_game | 81.0 | |
| summarize | 79.2 | |
| theory_of_mind | 78.8 | |
| javascript | 77.3 | |
| story_generation | 76.5 | |
| logic_with_navigation | 74.0 | |
| code_completion | 73.9 | |
| paraphrase | 73.3 | |
| code_generation | 71.8 | |
| simplify | 67.3 | |
| python | 60.0 | |
| plot_unscrambling | 58.6 | |
| typescript | 56.7 | |
| tablejoin | 50.1 |