Benchmark Almanac

AgentIF-OneDay

AutoClaw 79.9% tops its leaderboard ChatGPT agent 62.6% leads the models we track Not in model cards

Agents complete 104 tasks that stand for a full day of human work, life and study. Tasks include tedious workflows, inferring rules from examples and changing requirements.

First of the AgentIF series. An LLM judge script grades results. Manus, Genspark and ChatGPT Agent scored about 62 to 65% at launch. Not the THU-KEG AgentIF.

Made by xbench (HongShan) Released Dec 2025 General assistants All tasks public Last model added Jul 2025 Scores refreshed 2 Nov 2026
Level
Expert Specialists with years of training, often over hours of work
Scored
Graded by a model A language model scores each response, usually against a rubric.
Answers
Agent task Many steps in a terminal, browser, desktop, app, game or robot.
xbench-ai/AgentIF-OneDay 39 stars Paper Dataset Leaderboard xbench.org
5
Entries on its own leaderboard, 1 model we track

AutoClaw tops the leaderboard with 79.9%

Best score per entry from the benchmark's own leaderboard, read on 2 Nov 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is ChatGPT agent, at 62.6%.

  1. 79.9 AutoClaw
  2. 64.5 Manus
  3. 63.5 Genspark
  4. 62.6 ChatGPT agent
  5. 56.2 MiniMax-Agent
  1. 1 AutoClaw 79.9%
  2. 2 Manus 64.5%
  3. 3 Genspark 63.5%
  4. 4 ChatGPT agent 62.6%
  5. 5 MiniMax-Agent 56.2%
#EntryScoreSettingDateSource
1 AutoClaw 79.9% May 2026 Official leaderboard
2 Manus 64.5% Dec 2025 Official leaderboard
3 Genspark 63.5% Dec 2025 Official leaderboard
4 ChatGPT agent 62.6% Jul 2025 Official leaderboard
5 MiniMax-Agent 56.2% Dec 2025 Official leaderboard

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
0% 20% 40% 60% 80% 100% Near the ceiling (85%) ChatGPT agent 0% 20% 40% 60% 80% 100% Near the ceiling (85%) ChatGPT agent

No lab model card reports it yet

We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark.

More in Tool use

All 125 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. Muse Spark 1.1 88.1%
  • 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%
  • Claw-Eval In use 300 human-verified agent tasks (service orchestration, multimodal, multi-turn dialogue) audited via traces and snapshots. Rubric scores with Pass^3 consistency. Claude Sonnet 4.6 81.4%
  • DeepSearchQA In use Questions that need multi-step search and synthesis. Final answers graded against references. Kimi K3 95%