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 MiniMax-M3?
MiniMax-M3 is a model from MiniMax, released on Jun 1, 2026. Its SI-IQ is 108 (90% interval 102–114), estimated from 5 benchmark results by 2 independent evaluators.
Facts
- Developer
- MiniMax
- Released
- Jun 1, 2026
- Access
- Open weights (restricted licence)
- Benchmarks used
- 5 results from 2 evaluators
Ability by area (SI-IQ, 90% interval)
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 | Score | Evaluated by | Source |
|---|---|---|---|
| Chess Puzzles | 13.0% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| Mystery Game Puzzles | 8.0% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| OTIS Mock AIME 2024–2025 | 71.1% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| GPQA Diamond | 90.9% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| GPQA Diamond (OpenRouter run) | 91.5% | OpenRouter | OpenRouter benchmark datasets · data file |
| τ²-bench Airline (OpenRouter run) | 73.9% | OpenRouter | OpenRouter benchmark datasets · data file |
Human preference (Arena)
On the Arena leaderboard, where people vote between two anonymous answers, MiniMax-M3 has a rating of 1,441 (1,437–1,445) from 63,973 votes (published Oct 15, 2026). Human preference is a different measure from benchmark ability and is not part of SI-IQ. Arena (Hugging Face)
Models with similar SI-IQ
| Model | SI-IQ | |
|---|---|---|
| Gemini 2.5 Pro (Jun 2025)Google DeepMind | 108 | MiniMax-M3 vs Gemini 2.5 Pro (Jun 2025) |
| Gemini 3.5 Flash-LiteGoogle DeepMind | 109 | MiniMax-M3 vs Gemini 3.5 Flash-Lite |
| Claude Sonnet 4.5Anthropic | 110 | MiniMax-M3 vs Claude Sonnet 4.5 |
| gpt-oss-120bOpenAI | 106 | MiniMax-M3 vs gpt-oss-120b |
| Qwen 3.6 35B-A3BAlibaba | 111 | MiniMax-M3 vs Qwen 3.6 35B-A3B |
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); Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter..