Benchmark scores
Higher is better for all benchmark scores (SWE-bench, GPQA, HLE, Terminal-Bench, etc.). Green marks the model with the higher score.
LLM Comparison
MiniMax-M3 vs Composer 2.5: benchmark scores, pricing & comparison.
Side-by-side MiniMax-M3 vs Composer 2.5 comparison across SWE-bench, GPQA, HLE, Terminal-Bench, coding agent scores, token pricing, context window, and AskClash RWT. Green marks the winner on each benchmark.
Rank #9 vs #10AskClash overall scores 67.4 vs 67.0.
Pricing $0.30/$1.20 vs $0.50/$2.50Input and output token prices per 1M tokens when published.
Open Weight vs ProprietaryMiniMax vs Cursor.
MiniMax-M3 vs Composer 2.5 benchmark comparison
Green cells highlight the winning model for each metric. Scores are cached from the AskClash LLM leaderboard snapshot.
| Metric | MiniMax-M3 | Composer 2.5 |
|---|---|---|
| Overall Score | 67.4 | 67.0 |
| Leaderboard Rank | #9 | #10 |
| RWT | 9.5 | 8.5 |
| Coding Agent Index | — | 51.8 |
| HLE | 37.1 | — |
| GPQA | 92.9 | — |
| IFEval | 82.9 | — |
| SWE-bench | 80.5 | — |
| SWE-Pro | 59.0 | 47.0 |
| SWE-Atlas | — | 72.0 |
| Terminal-Bench | 66.0 | 69.3 |
| OSWorld | 70.1 | — |
| MCP Atlas | 74.2 | — |
| Finance Agent | 48.3 | — |
| MMMU-Pro | 78.1 | — |
| Tau2 | 88.9 | — |
| Input Price (per 1M tokens) | $0.30 | $0.50 |
| Output Price (per 1M tokens) | $1.20 | $2.50 |
| Context Window | 1M | 200K |
| Benchmark Cells | 13 | 4 |
More MiniMax-M3 and Composer 2.5 comparisons
Explore how MiniMax-M3 and Composer 2.5 stack up against other top-ranked LLMs.
Claude Mythos/Fable 5 vs MiniMax-M3Claude Mythos/Fable 5 vs Composer 2.5Claude Opus 4.8 (Adaptive) vs MiniMax-M3Claude Opus 4.8 (Adaptive) vs Composer 2.5GPT-5.5 xHigh vs MiniMax-M3GPT-5.5 xHigh vs Composer 2.5GLM-5.2 vs MiniMax-M3GLM-5.2 vs Composer 2.5Claude Opus 4.7 (Adaptive) vs MiniMax-M3Claude Opus 4.7 (Adaptive) vs Composer 2.5
How to read this comparison
Token pricing
Lower is better for input and output prices. Green marks the cheaper model per 1M tokens.
Coverage matters
Models with fewer disclosed benchmark cells may have inflated percentile scores. Check the benchmark cell count for context.
This comparison page is generated from the AskClash LLM leaderboard cache. Open the live leaderboard for real-time scores and interactive filtering.