#37MiniMax· released 2026-05-31

MiniMax: MiniMax M3.

MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,...

View on OpenRouter ↗
Context
1.0M
Input $/1M
$0.30
Output $/1M
$1.20
Modality
text · image · video
Arena Elo
1535
LMArena — human blind votes
AA Index
59.0
Artificial Analysis composite
LiveBench
71.3
Contamination-resistant
GPQA
80.8%
Grad-level science reasoning
MMLU-Pro
81.6%
Broad knowledge
SWE-bench
65.1%
Real GitHub bugs (Verified)
Terminal-Bench
53.2%
Agentic shell tasks
ARC-AGI
12.1%
Abstract reasoning
Aider Polyglot
71.9%
Multi-language code editing
Speed
96 tok/s
Median output speed

Profile

MiniMax: MiniMax M3 is MiniMax's entry at position #37 on the current aggregated leaderboard. Its strongest signals are graduate-level science reasoning, real-world software engineering, unattended agentic work, and very long context. Every figure below is aggregated from the public sources listed on the homepage and refreshed on each pipeline run.

The score spread is comparatively complete for this model, but the standing advice still applies: treat gaps of a few points as ties and validate on your own workload before committing.

How this model's position is computed is documented in the methodology; what each benchmark actually measures is in the benchmark guide. See a figure that disagrees with its source? Report it — accepted corrections apply on the next refresh.

Nearby on the leaderboard