BlogLLM Benchmark Scores as Published, Side by Side
MM-ClawBench: Results as Published
Every result the catalog holds for MM-ClawBench: 6 models from 4 developers, read from 1 page. Each number is as printed by the company named beside it; nothing is ranked or rescored. Models are listed by developer, newest first.
Area Tool Use and WorkflowsMaintainer MiniMaxIntroduced 2026Data version 2026103001, as of 2026-10-30
MM Claw is an evaluation set MiniMax built from tasks people commonly give OpenClaw agents. It covers work and life jobs, including personal learning plans, office document processing and delivery, scheduled research and investment advice, and code development and maintenance. The end-to-end score on that set is accuracy. MiniMax also reports a skill-compliance rate on the same name for 40 complex skills, each longer than 2,000 tokens.
MM-ClawBench
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.4OpenAI, released 2026-03-05 | 73.6% | MiniMax, MiniMax-M2.7 model cardchecked 2026-10-15 | - |
| Claude Sonnet 4.6Anthropic, released 2026-02-17 | 64.2% | MiniMax, MiniMax-M2.7 model cardchecked 2026-10-15 | - |
| Claude Opus 4.6Anthropic, released 2026-02-05 | 75.4% | MiniMax, MiniMax-M2.7 model cardchecked 2026-10-15 | - |
| Gemini 3.1 Pro PreviewGoogle, released 2026-02-19 | 61.8% | MiniMax, MiniMax-M2.7 model cardchecked 2026-10-15 | - |
| MiniMax M2.7MiniMax, released 2026-03-18 | 62.7% | MiniMax, MiniMax-M2.7 model cardchecked 2026-10-15 | - |
| MiniMax M2.5MiniMax, released 2026-02-12 | 57.6% | MiniMax, MiniMax-M2.7 model cardchecked 2026-10-15 | - |
Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.