IQ of SI
English ▾

Automated page. Written and translated automatically by AI and checked against the sources by software; no human editor. SI-IQ is an index built from public benchmark results, not an IQ test. How it works

How smart is Qwen3-Max?

106 SI-IQ · 90% interval 103–110
Not ranked: too little data

Qwen3-Max is a model from Alibaba, released on Sep 24, 2025. Its SI-IQ is 106 (90% interval 103–110), estimated from 6 benchmark results by 1 independent evaluators. Its strongest area is Math and its weakest is Knowledge.

This model has fewer than 5 independent benchmark results or fewer than 2 different evaluators, so it is shown but not ranked.

Facts

Developer
Alibaba
Released
Sep 24, 2025
Access
API access
Benchmarks used
6 results from 1 evaluators

Ability by area (SI-IQ, 90% interval)

6080100120140160Math107Knowledge105

Every benchmark result we used

Each score is one value exactly as the source published it: for Epoch AI the best of its run settings, for OpenRouter the median hosting provider (the developer's own hosting when it ran the test). “Implied SI-IQ” is the ability that this single result points to; the overall SI-IQ is their precision-weighted average.

Benchmark Area Score Implied SI-IQ Evaluated by As listed by the source Retrieved Source
Mystery Game Puzzles Reasoning 5.0% 70 Epoch AI qwen3-max-2025-09-23 Oct 18, 2026 Epoch AI Benchmarking Hub · data file
FrontierMath Tiers 1–3 Math 18.9% 105 Epoch AI qwen3-max-2025-09-23 Oct 18, 2026 Epoch AI Benchmarking Hub · data file
MATH Level 5 Math 97.1% 111 Epoch AI qwen3-max-2025-09-23 Oct 18, 2026 Epoch AI Benchmarking Hub · data file
OTIS Mock AIME 2024–2025 Math 73.3% 104 Epoch AI qwen3-max-2025-09-23 Oct 18, 2026 Epoch AI Benchmarking Hub · data file
GPQA Diamond Knowledge 72.6% 98 Epoch AI qwen3-max-2025-09-23 Oct 18, 2026 Epoch AI Benchmarking Hub · data file
SimpleQA Verified Knowledge 48.7% 128 Epoch AI qwen3-max-2025-09-23 Oct 18, 2026 Epoch AI Benchmarking Hub · data file

Human preference (Arena)

On the Arena leaderboard, where people vote between two anonymous answers, Qwen3-Max has a rating of 1,423 (1,417–1,430) from 9,027 votes (published Oct 15, 2026). Human preference is a different measure from benchmark ability and is not part of SI-IQ. Arena (Hugging Face)

Computed with SI-IQ method version 1.0.0. Read the full methodology

Data sources: Arena leaderboard dataset on Hugging Face (CC BY 4.0); Epoch AI, ‘Capabilities & benchmarking’ (CC BY 4.0).