MiniMax-M2.5
Prior MiniMax coding model for agent workflows, office edits, and automation
Shogo Agent Index
Estimated#74 of 9228.4range 19.5–37.2 · 19% of weight measured
Agentic coding
32
23% measured
Long-horizon autonomy
34
30% measured
Knowledge-work agents
24
estimated
Computer use
17
estimated
Security & context learning
27
50% measured
Agentic benchmarks
| Benchmark | Score | Normalised | Index weight | Harness / source |
|---|---|---|---|---|
| SWE-bench Verified | est. | 8% | predicted 63 | |
| Terminal-Bench 2.0 | 42.7% | 8% | cchuter | |
| DeepSWE | est. | 6% | predicted 21 | |
| FrontierSWE | est. | 5% | predicted 3 | |
| CursorBench | est. | 4% | predicted 25 | |
| GSO-Bench | est. | 4% | predicted 7 | |
| METR Time Horizons | est. | 10% | predicted 62 | |
| Vending-Bench 2 | $-23 | 6% | source | |
| PostTrainBench | est. | 4% | predicted 16 | |
| APEX-Agents | est. | 8% | predicted 22 | |
| GDPval | est. | 7% | predicted 28 | |
| Remote Labor Index | est. | 5% | predicted 1 | |
| DeepResearch Bench | est. | 5% | predicted 44 | |
| OSWorld 2.0 | est. | 6% | predicted 2 | |
| OSWorld | est. | 4% | predicted 39 | |
| Cybench | est. | 5% | predicted 43 | |
| CL-bench | 11.4% | 5% | source |
Other benchmarks
| Benchmark | Category | Score | Source |
|---|---|---|---|
| ARC-AGI | Reasoning | 63.7% | source |
| ARC-AGI-2 | Reasoning | 4.9% | source |
| ProofBench | Math | 4.0% | source |
| WebDev Arena | Coding | 1384 | source |
API hosts
Live prices and uptime for every host serving MiniMax-M2.5 via OpenRouter.
| Host | Input | Output | Cache read | Agent $/M | Uptime 24h | Quant |
|---|---|---|---|---|---|---|
| Venice | $0.27 | $0.95 | $0.030 | $0.11 | 99.8% | — |
| GMICloud | $0.30 | $1.20 | $0.030 | $0.12 | 98.5% | fp8 |
| AtlasCloud | $0.29 | $1.20 | $0.060 | $0.15 | 100.0% | fp8 |
| DigitalOcean | $0.30 | $1.20 | $0.060 | $0.15 | 98.5% | — |