← leaderboard
MiniMax M3
SUSPICIOUSMiniMax · minimax/minimax-m3 · verdict as of 2026-08-10
Flags
- · instruction-following: score 75.0 vs baseline 98.3 (z=-11.64, drop 23.7%)
- · game-design: score 33.1 vs baseline 46.0 (z=-3.24, drop 28.2%)
Daily test history (90 days, baseline band = trailing mean ± 2σ)
Math & Reasoning
NOT COOKED
Code Generation
NOT COOKED
Instruction Following
SUSPICIOUS
Report Analysis
NOT COOKED
Web Design
NOT COOKED
Structured Extraction
NOT COOKED
Game Design
SUSPICIOUS
Creative Writing
no data
Summarization
NOT COOKED
Customer Service
NOT COOKED
Serving providers (last 30 days)
striped = multiple providers served this model that day (hover for detail)
Public benchmarks overall 74.6
MMLU-Pro
—
GPQA Diamond
93
SWE-bench Verified
—
LMArena Elo
1447
AIME 2025
—
retrieved 2026-07-03 from public sources — see methodology
Recent samples (latest run, one per test case)
No samples from the latest run.