Benchmark Almanac

Collab-Overcooked

Claude Sonnet 4 86.7% leads since May 2025
Not in model cards

Two agents exchange messages and use different kitchen equipment to fill orders. The benchmark measures task completion and progress through required cooking steps.

The authors tested 13 language models on 30 tasks. An independent team ran levels 1, 2, 5, and 6 to study partner changes.

Made by
Haochen Sun and colleagues
Released
February 2025
Kind
Multi-agent games
Level
Everyday
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text, Actions
Interaction
Multi-agent
Environment
Overcooked
Items from
Written by experts
Refresh
One fixed set
Run by
Anyone
In views
Games, Agents
Strongest model tested
o4-mini #95 on the capability index
Last model added
May 2025
Scores refreshed
November 20th, 2026
13
Entries in its paper, 5 models we track

Claude Sonnet 4 leads with 86.7% among models released by May 2025 Claude Sonnet 4 leads with 86.7% among models released by May 2025

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The paper gives no overall score, so this is the plain mean of success rates at its six complexity levels. 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. This view shows only models we track, each at its best.

The newest model with a score here came out in May 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Claude Sonnet 4 86.7%
  2. 2 o4-mini 81.7%
  3. 3 DeepSeek-R1 (snapshot unspecified) 75.7%
  4. 4 GPT-4o (snapshot unspecified) 48%
  5. 5 DeepSeek-V3 (snapshot unspecified) 42%
  6. 6 Qwen2.5-72B 27.3%
  7. 7 Llama-3.1-70B-Instruct 23.3%
  8. 8 Qwen2.5-32B-Instruct 21.3%
  9. 9 o1-mini 12%
  10. 10 GPT-3.5 Turbo Jan 2024 8.33%
  11. 11 Qwen2.5-14B-Instruct 6.00%
  12. 12 Qwen2.5-7B-Instruct 1.33%
  13. 13 Llama-3.1-8B-Instruct 0.67%

Source: The Collab-Overcooked paper

#EntryScoreSettingDateSource
1 Claude Sonnet 4 86.7% May 2025 Official leaderboard
2 o4-mini 81.7% Apr 2025 Official leaderboard
3 DeepSeek-R1 (snapshot unspecified) 75.7% Official leaderboard
4 GPT-4o (snapshot unspecified) 48% Official leaderboard
5 DeepSeek-V3 (snapshot unspecified) 42% Official leaderboard
6 Qwen2.5-72B 27.3% Sep 2024 Official leaderboard
7 Llama-3.1-70B-Instruct 23.3% Official leaderboard
8 Qwen2.5-32B-Instruct 21.3% Official leaderboard
9 o1-mini 12% Sep 2024 Official leaderboard
10 GPT-3.5 Turbo Jan 2024 8.33% Jan 2024 Official leaderboard
11 Qwen2.5-14B-Instruct 6.00% Official leaderboard
12 Qwen2.5-7B-Instruct 1.33% Official leaderboard
13 Llama-3.1-8B-Instruct 0.67% Official leaderboard
#ModelScoreSettingDateSource
1 Claude Sonnet 4 86.7% May 2025 Official leaderboard
2 o4-mini 81.7% Apr 2025 Official leaderboard
3 Qwen2.5-72B 27.3% Sep 2024 Official leaderboard
4 o1-mini 12% Sep 2024 Official leaderboard
5 GPT-3.5 Turbo Jan 2024 8.33% Jan 2024 Official leaderboard

3 models have held number one in the benchmark’s paper since its release in February 2025

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

Qwen OpenAI Anthropic
Benchmark paper Passed 85% in May 2025, by Claude Sonnet 4

Source: The Collab-Overcooked paper

Claude Sonnet 4 on the official leaderboard passed 85% in May 2025

That was 3 months after its release. By the same age, 12% of the benchmarks from 2025 had passed 85%. All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
In the benchmark’s paper Best so far
0% 20% 40% 60% 80% 100% 2024 2025 Near the ceiling (85%) Claude Sonnet 4 o4-mini Qwen2.5-72B o1-mini GPT-3.5 Turbo 0% 20% 40% 60% 80% 100% 2024 2025 Near the ceiling (85%) Claude Sonnet 4 Qwen2.5-72B

Source: Official Collab-Overcooked 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 Games and long tasks

All 261 in Games and long tasks
  • ARC-AGI-3 In use Interactive game environments where agents must explore, infer goals and rules, and act efficiently without instructions. GPT-6.2 Astra 100%
  • Orak Fading LLM agents play 12 popular video games across most genres through an MCP interface. Each game scores in-game progress or outcomes, and the board ranks models by their average rank across the games. Gemini 2.5 Pro 3.90
  • MineDojo Fading Agents complete open-ended tasks in Minecraft, such as mining blocks and fighting mobs, from a large simulation suite. Programmatic checks score many tasks. Seed2.0 Pro 49%
  • TextQuests Only in system cards Agents play 25 classic interactive fiction games. The score is harmful actions per 100 steps, where lower is better. The board also reports game progress. Llama 4 Scout 7.00
  • GameWorld Only in system cards Multimodal game agents play 34 browser games across 170 tasks. Outcomes are checked from the game state. GPT-6.1 Terra 76.8%
  • Swarm Bench Only in system cards Models orchestrate agent swarms on complex tasks that benefit from coordinated decomposition and parallel execution. Kimi K3 76.3%

Data audit

All 8 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the benchmark’s paper (13 entries).
  • Pass Models scored 5 models we track, plus 8 other entries on its leaderboard.
  • Note Newest score The newest came 18 months ago, in May 2025, for Claude Sonnet 4. That’s expected for a saturated test.
  • Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is o4-mini (#95).
  • Note Headroom Saturated. The best is 86.7%, past 85% since May 2025.
Leaderboard import 3 pass
  • Pass Importer We read the board on aclanthology.org: 5 rows for models we track and 8 other entries 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