Across 10 shared benchmarks, MiniMax M2.5 leads overall: MiniMax M2.5 wins 8, M2.1 wins 1, with 1 ties and an average score difference of +8.21.
MiniMaxAI · 2026-02-12 · Reasoning model
MiniMaxAI · 2025-12-23 · AI model
Grouped by capability, sorted by largest gap within each. 10 shared benchmarks.
| Benchmark | MiniMax M2.5 | M2.1 | Diff |
|---|---|---|---|
| SWE-Bench Pro - Public | 55.4013 / 36thinking + 使用工具 | 32.6035 / 36thinking + 使用工具 | +22.80 |
| SWE-bench Verified | 80.209 / 103thinking + 使用工具 | 74.8030 / 103thinking | +5.40 |
| Benchmark | MiniMax M2.5 | M2.1 | Diff |
|---|---|---|---|
| GPQA Diamond | 85.2045 / 175Thinking (No Tools) | 8166 / 175thinking | +4.20 |
| HLE | 19.4098 / 149Thinking (No Tools) | 2286 / 149thinking | -2.60 |
| Benchmark | MiniMax M2.5 | M2.1 | Diff |
|---|---|---|---|
| τ²-Bench - Telecom | 97.8010 / 35thinking + 使用工具 | 8722 / 35thinking + 使用工具 | +10.80 |
| Benchmark | MiniMax M2.5 | M2.1 | Diff |
|---|---|---|---|
| BrowseComp | 76.3016 / 43thinking + 使用工具 | 47.4035 / 43thinking + 使用工具 | +28.90 |
| Benchmark | MiniMax M2.5 | M2.1 | Diff |
|---|---|---|---|
| Terminal Bench 2.0 | 51.7027 / 43thinking + 使用工具 | 47.9032 / 43thinking + 使用工具 | +3.80 |
| Benchmark | MiniMax M2.5 | M2.1 | Diff |
|---|---|---|---|
| Pinch Bench | 87.806 / 37Thinking (With Tools) | 84.3018 / 37Thinking (With Tools) | +3.50 |
| Benchmark | MiniMax M2.5 | M2.1 | Diff |
|---|---|---|---|
| IF Bench | 7010 / 27thinking + 使用工具 | 7010 / 27thinking + 使用工具 | — |
| Benchmark | MiniMax M2.5 | M2.1 | Diff |
|---|---|---|---|
| AIME2025 | 86.3048 / 106Thinking (No Tools) | 8156 / 106thinking | +5.30 |
| Field | MiniMax M2.5 | M2.1 |
|---|---|---|
| Publisher | MiniMaxAI | MiniMaxAI |
| Release date | 2026-02-12 | 2025-12-23 |
| Model type | Reasoning model | AI model |
| Architecture | MoE | MoE |
| Parameters | 2290.0 | 2300.0 |
| Context length | 128K | 200K |
| Max output | Not available | 131072 |
Prices use DataLearner records when available; missing fields are not inferred.
| Item | MiniMax M2.5 | M2.1 |
|---|---|---|
| Text input | $0.3 / 1M tokens | 0.3 美元/100 万tokens |
| Text output | $2.4 / 1M tokens | 1.2 美元/100 万tokens |
| Cache read | Not public | 0.03 美元/100 万tokens |
| Cache write | Not public | 0.375 美元/100 万tokens |
On average across the 10 shared benchmarks, MiniMax M2.5 scores 8.21 higher.
Largest single-benchmark gap: BrowseComp — MiniMax M2.5 76.30 vs M2.1 47.40 (+28.90).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.
