Benchmark Almanac

AgentCLUE-Task

Kimi K3 49.5 leads since July 2026
Not in model cards

Agents carry out interactive tasks in Chinese.

SuperCLUE reports scores in a public workbook.

Made by
SuperCLUE
Kind
Apps and MCP servers
Level
Everyday
Scored
Settled by events
Answers
Agent task
Measures
Capability
Language
Chinese
In and out
Text → Actions
Interaction
Agentic
Environment
APIs and tools
Run by
A host
Based on
AgentCLUE
In views
Agents
Strongest model tested
GPT-5.5 #21 on the capability index
Last model added
July 2026
Scores refreshed
November 20th, 2026
6
Models scored on its own leaderboard

Kimi K3 leads with 49.5 among models released by July 2026

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.

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

  1. 1 Kimi K3 max 49.5
  2. 2 GPT-5.5 high 47.5
  3. 3 Claude Opus 4.7 high 44.5
  4. 4 Gemini 3.1 Pro high 43.6
  5. 5 GLM-5.2 max 37.6
  6. 6 DeepSeek-V4-Pro max 34.6

Source: Official AgentCLUE-Task leaderboard

#ModelScoreSettingDateSource
1 Kimi K3 49.5 max Jul 2026 Official leaderboard
2 GPT-5.5 47.5 high Apr 2026 Official leaderboard
3 Claude Opus 4.7 44.5 high Apr 2026 Official leaderboard
4 Gemini 3.1 Pro 43.6 high Feb 2026 Official leaderboard
5 GLM-5.2 37.6 max Jun 2026 Official leaderboard
6 DeepSeek-V4-Pro 34.6 max Apr 2026 Official leaderboard

Rankings by part

We read 5 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as 外卖点单, 电商零售 and 电子政务. Gemini 3.1 Pro leads 2 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

外卖点单

6 models we track have a score here.

  1. 1 GPT-5.5 ↑1 63.2
  2. 2 Kimi K3 ↓1 57.9
  3. 3 Gemini 3.1 Pro ↑1 42.1
  4. 4 Claude Opus 4.7 ↓1 42.1
  5. 5 DeepSeek-V4-Pro ↑1 31.6
  6. 6 GLM-5.2 ↓1 31.6

4 models have held number one on the official leaderboard since 2023

Lab claims count from the day the card came out, independent runs from the model’s release.

Google DeepMind Anthropic OpenAI Moonshot AI
Own leaderboard Not tested since Jul 2026
Kimi K3 49.5

Source: Official AgentCLUE-Task leaderboard

Scores over time

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

Plot with other benchmarks
On the benchmark’s own leaderboard
30.0 35.0 40.0 45.0 50.0 2026 30.0 35.0 40.0 45.0 50.0 2026

Source: Official AgentCLUE-Task leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Tool use

All 265 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 Neon 56.3%
  • 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%

Data audit

8 of 9 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark

Catalog record 1 warning 1 pass
  • Warning Task access We don’t know whether the tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (6 entries).
  • Pass Models scored 6 models we track.
  • Pass Newest score The newest is from July 2026, for Kimi K3.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.5 (#21).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on superclueai.com: 6 rows for models we track at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check