Benchmark Almanac

MM-ClawBench

Claude Opus 4.6 75.4% leads since March 2026
Fading

Agents complete common OpenClaw tasks such as study planning, office documents, scheduled research, and code maintenance. Scoring is accuracy.

In-house: MiniMax never released the tasks or explained the scoring.

Made by
MiniMax
Released
March 2026
Kind
General assistants
Level
Everyday
Scored
Graded by a model
Answers
Agent task
Tasks
Kept private
Strongest model tested
GPT-5.4 #27 on the capability index
Last in a launch post
March 2026, dropped before it saturated
Last model added
May 2026
Scores refreshed
November 12th, 2026
6
Models scored on its own leaderboard
1
Launch post lists it, from 1 lab
2
System cards report it, from 1 lab

Claude Opus 4.6 leads with 75.4% among models released by March 2026

Best score per model from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in March 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 75.4 Claude Opus 4.6
  2. 73.6 GPT-5.4
  3. 64.2 Claude Sonnet 4.6
  4. 62.7 MiniMax-M2.7
  5. 61.8 Gemini 3.1 Pro
  6. 57.6 MiniMax-M2.5
  1. 1 Claude Opus 4.6 75.4%
  2. 2 GPT-5.4 73.6%
  3. 3 Claude Sonnet 4.6 64.2%
  4. 4 MiniMax-M2.7 62.7%
  5. 5 Gemini 3.1 Pro 61.8%
  6. 6 MiniMax-M2.5 57.6%

Source: Official MM-ClawBench leaderboard

#ModelScoreSettingDateSource
1 Claude Opus 4.6 75.4% Feb 2026 Official leaderboard
2 GPT-5.4 73.6% Mar 2026 Official leaderboard
3 Claude Sonnet 4.6 64.2% Feb 2026 Official leaderboard
4 MiniMax-M2.7 62.7% Mar 2026 Official leaderboard
5 Gemini 3.1 Pro 61.8% Feb 2026 Official leaderboard
6 MiniMax-M2.5 57.6% Feb 2026 Official leaderboard

Labs report scores for 2 models in their cards. We plot every source in the charts below.

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release. We draw one row each for the benchmark’s own leaderboard and what labs claim in their cards.

Anthropic MiniMax
Own leaderboard Not tested since Mar 2026
Claude Opus 4.6 75.4%
Lab claims Not tested since May 2026
MiniMax-M2.7 62.7%

Sources: Official MM-ClawBench leaderboard and labs’ model cards

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard Reported by the lab in its model card
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 4.6 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 4.6

Sources: Official MM-ClawBench leaderboard and labs’ model cards

Labs headline it less often now, last in March 2026

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

MiniMax
DateLaunch postScoreAs the post writes it
Mar 18, 2026 MiniMax-M2.7 62.7 MM-ClawBench (MM Claw)

2 system cards report it, from 1 lab

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
May 26, 2026 MiniMax-M2.7 technical report 62.7 MM Claw (Table 4) Capabilities
May 26, 2026 MiniMax-M2.5 technical report 57.6 MM Claw (Table 4) Capabilities

2 lab claims next to the leaderboard

Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. 2 of them match, which usually means one side copied the other. Every benchmark.

Model Lab says Board says Gap
MiniMax-M2.7 62.7% MM-ClawBench (MM Claw) 62.7% Leaderboard Same
MiniMax-M2.5 57.6% MM Claw (Table 4) 57.6% Leaderboard Same

More in Tool use

All 144 in Tool use
  • AutomationBench In use Agent completes end-to-end business workflows inside a simulated company exposing dozens of REST APIs. Graded on resulting state. Gemini 4 Carbon 53.9%
  • MCP Atlas In use Multi-step workflows over real MCP servers. Scores tool-use competency on realistic tasks. Gemini 4 Argon 90.7%
  • BFCL V4 In use Models call functions and tools, including multi-turn agentic tasks with web search and memory. Checked against expected calls and outcomes. Qwen3.7-Max 75%
  • τ³-bench Banking In use Agents act as banking customer-service reps over many turns with a simulated user, following policy. Pass^1 on the final database state. Qwen 3.8 Max 55.1%
  • Toolathlon-Verified In use 108 long-horizon tasks across 32 apps through about 600 MCP tools. Execution-based checkers inspect final environment state. GLM-5.3-Flash 78.4%
  • DeepSearchQA In use Questions that need multi-step search and synthesis. Final answers graded against references. Kimi K3 95%