Hanabi Agent Benchmark
Models play cooperative Hanabi in four-player teams. The benchmark scores shared board progress while hints and lives constrain coordination.
The October 2026 dataset compares 21 baseline model configurations. Claude Opus 5.5 leads with mean progress 20.50 out of 25 across four seeds.
- Made by
- flatkey
- Released
- October 2026
- Kind
- Multi-agent games
- Level
- Everyday
- Scored
- Settled by events
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Multi-agent
- Environment
- Hanabi
- Items from
- Live play or votes
- Refresh
- Live
- Run by
- Anyone
- In views
- Games, Agents, Kept fresh
- Strongest model tested
- GPT-6 Astra #4 on the capability index
- Last model added
- October 2026
- Scores refreshed
- November 20th, 2026
Claude Opus 5.5 leads with 20.5 Claude Opus 5.5 leads with 20.5
Best score per entry from the benchmark's own leaderboard, read on November 20th, 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. This view shows only models we track, each at its best.
- 20.5 Claude Opus 5.5
- 18.8 GPT-6 Astra
- 18.0
Gemini 3.1 Pro
- 18.0 Claude Fable 5.1
- 16.3 GPT-6 Sol
- 16.0
Qwen 3.8 Max
- 16.0
Gemini 3.8 Flash
- 15.5 Seed 2.1 Lite
- 15.3
DeepSeek-V4-Pro
- 14.3 DeepSeek Flash
- 13.3
Tencent Hy3
- 12.8 Claude Haiku 5.5
- 12.5
Seed2.1 Pro
- 11.3 Kimi K3
- 10.5 Claude Sonnet 5.5
- 9.50
Qwen3.8 Flash
- 9.50 MiMo-V2.6 Flash
- 7.25 MiMo-V2.6 Pro
- 6.75 GPT-6 Luna
- 5.75 GLM-5.3-Flash
- 4.50 GLM-5.3
- 1 Claude Opus 5.5 Low 20.5
- 2 GPT-6 Astra Low 18.8
- 3
Gemini 3.1 Pro Low 18.0
- 4 Claude Fable 5.1 Low 18.0
- 5 GPT-6 Sol Low 16.3
- 6
Qwen 3.8 Max Single 16.0
- 7
Gemini 3.8 Flash Low 16.0
- 8 Seed 2.1 Lite Single 15.5
- 9
DeepSeek-V4-Pro Single 15.3
- 10 DeepSeek Flash Low 14.3
- 11
Tencent Hy3 Low 13.3
- 12 Claude Haiku 5.5 Low 12.8
- 13
Seed2.1 Pro Single 12.5
- 14 Kimi K3 Low 11.3
- 15 Claude Sonnet 5.5 Low 10.5
- 16
Qwen3.8 Flash Single 9.50
- 17 MiMo-V2.6 Flash Single 9.50
- 18 MiMo-V2.6 Pro Single 7.25
- 19 GPT-6 Luna Low 6.75
- 20 GLM-5.3-Flash Low 5.75
- 21 GLM-5.3 Low 4.50
Source: Official Hanabi Agent Benchmark leaderboard
- 20.5 Claude Opus 5.5
- 18.8 GPT-6 Astra
- 18.0
Gemini 3.1 Pro
- 18.0 Claude Fable 5.1
- 16.3 GPT-6 Sol
- 16.0
Qwen 3.8 Max
- 16.0
Gemini 3.8 Flash
- 15.3
DeepSeek-V4-Pro
- 13.3
Tencent Hy3
- 12.8 Claude Haiku 5.5
- 12.5
Seed2.1 Pro
- 11.3 Kimi K3
- 10.5 Claude Sonnet 5.5
- 9.50
Qwen3.8 Flash
- 9.50 MiMo-V2.6 Flash
- 7.25 MiMo-V2.6 Pro
- 6.75 GPT-6 Luna
- 5.75 GLM-5.3-Flash
- 4.50 GLM-5.3
- 1 Claude Opus 5.5 Low 20.5
- 2 GPT-6 Astra Low 18.8
- 3
Gemini 3.1 Pro Low 18.0
- 4 Claude Fable 5.1 Low 18.0
- 5 GPT-6 Sol Low 16.3
- 6
Qwen 3.8 Max Single 16.0
- 7
Gemini 3.8 Flash Low 16.0
- 8
DeepSeek-V4-Pro Single 15.3
- 9
Tencent Hy3 Low 13.3
- 10 Claude Haiku 5.5 Low 12.8
- 11
Seed2.1 Pro Single 12.5
- 12 Kimi K3 Low 11.3
- 13 Claude Sonnet 5.5 Low 10.5
- 14
Qwen3.8 Flash Single 9.50
- 15 MiMo-V2.6 Flash Single 9.50
- 16 MiMo-V2.6 Pro Single 7.25
- 17 GPT-6 Luna Low 6.75
- 18 GLM-5.3-Flash Low 5.75
- 19 GLM-5.3 Low 4.50
Source: Official Hanabi Agent Benchmark leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 | 20.5 | Low | Sep 2026 | Official leaderboard |
| 2 | GPT-6 Astra | 18.8 | Low | Sep 2026 | Official leaderboard |
| 3 | | 18.0 | Low | Feb 2026 | Official leaderboard |
| 4 | Claude Fable 5.1 | 18.0 | Low | Sep 2026 | Official leaderboard |
| 5 | GPT-6 Sol | 16.3 | Low | Sep 2026 | Official leaderboard |
| 6 | | 16.0 | Single | Aug 2026 | Official leaderboard |
| 7 | | 16.0 | Low | Sep 2026 | Official leaderboard |
| 8 | Seed 2.1 Lite | 15.5 | Single | Official leaderboard | |
| 9 | | 15.3 | Single | Apr 2026 | Official leaderboard |
| 10 | DeepSeek Flash | 14.3 | Low | Official leaderboard | |
| 11 | | 13.3 | Low | Jul 2026 | Official leaderboard |
| 12 | Claude Haiku 5.5 | 12.8 | Low | Oct 2026 | Official leaderboard |
| 13 | | 12.5 | Single | Jun 2026 | Official leaderboard |
| 14 | Kimi K3 | 11.3 | Low | Jul 2026 | Official leaderboard |
| 15 | Claude Sonnet 5.5 | 10.5 | Low | Sep 2026 | Official leaderboard |
| 16 | | 9.50 | Single | Aug 2026 | Official leaderboard |
| 17 | MiMo-V2.6 Flash | 9.50 | Single | Sep 2026 | Official leaderboard |
| 18 | MiMo-V2.6 Pro | 7.25 | Single | Sep 2026 | Official leaderboard |
| 19 | GPT-6 Luna | 6.75 | Low | Sep 2026 | Official leaderboard |
| 20 | GLM-5.3-Flash | 5.75 | Low | Aug 2026 | Official leaderboard |
| 21 | GLM-5.3 | 4.50 | Low | Aug 2026 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 | 20.5 | Low | Sep 2026 | Official leaderboard |
| 2 | GPT-6 Astra | 18.8 | Low | Sep 2026 | Official leaderboard |
| 3 | | 18.0 | Low | Feb 2026 | Official leaderboard |
| 4 | Claude Fable 5.1 | 18.0 | Low | Sep 2026 | Official leaderboard |
| 5 | GPT-6 Sol | 16.3 | Low | Sep 2026 | Official leaderboard |
| 6 | | 16.0 | Single | Aug 2026 | Official leaderboard |
| 7 | | 16.0 | Low | Sep 2026 | Official leaderboard |
| 8 | | 15.3 | Single | Apr 2026 | Official leaderboard |
| 9 | | 13.3 | Low | Jul 2026 | Official leaderboard |
| 10 | Claude Haiku 5.5 | 12.8 | Low | Oct 2026 | Official leaderboard |
| 11 | | 12.5 | Single | Jun 2026 | Official leaderboard |
| 12 | Kimi K3 | 11.3 | Low | Jul 2026 | Official leaderboard |
| 13 | Claude Sonnet 5.5 | 10.5 | Low | Sep 2026 | Official leaderboard |
| 14 | | 9.50 | Single | Aug 2026 | Official leaderboard |
| 15 | MiMo-V2.6 Flash | 9.50 | Single | Sep 2026 | Official leaderboard |
| 16 | MiMo-V2.6 Pro | 7.25 | Single | Sep 2026 | Official leaderboard |
| 17 | GPT-6 Luna | 6.75 | Low | Sep 2026 | Official leaderboard |
| 18 | GLM-5.3-Flash | 5.75 | Low | Aug 2026 | Official leaderboard |
| 19 | GLM-5.3 | 4.50 | Low | Aug 2026 | Official leaderboard |
Rankings by part
We read 4 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Seed 101, Seed 102 and Seed 103. Claude Opus 5.5 leads 3 of them, and Claude Fable 5.1 leads the other one. Arrows show how far a model moved from its overall rank among the models we track.
Seed 101
19 models we track have a score here.
- 1 Claude Opus 5.5 20.0
- 2 Claude Fable 5.1 ↑2 18.0
- 3 GPT-6 Astra ↓1 18.0
- 4
Gemini 3.1 Pro ↓1 17.0
- 5
DeepSeek-V4-Pro ↑3 17.0
- 6
Gemini 3.8 Flash ↑1 17.0
- 7 GPT-6 Sol ↓2 17.0
- 8
Seed2.1 Pro ↑3 15.0
- 9
Qwen 3.8 Max ↓3 15.0
- 10
Tencent Hy3 ↓1 12.0
Seed 102
19 models we track have a score here.
- 1 Claude Opus 5.5 21.0
- 2
Gemini 3.1 Pro ↑1 20.0
- 3 GPT-6 Astra ↓1 20.0
- 4 Claude Fable 5.1 17.0
- 5
Qwen 3.8 Max ↑1 16.0
- 6
Tencent Hy3 ↑3 15.0
- 7
Gemini 3.8 Flash 15.0
- 8
DeepSeek-V4-Pro 14.0
- 9 GPT-6 Sol ↓4 14.0
- 10
Seed2.1 Pro ↑1 13.0
Seed 103
19 models we track have a score here.
- 1 Claude Fable 5.1 ↑3 21.0
- 2 GPT-6 Astra 19.0
- 3 Claude Opus 5.5 ↓2 19.0
- 4 Kimi K3 ↑8 17.0
- 5
Gemini 3.8 Flash ↑2 17.0
- 6 GPT-6 Sol ↓1 17.0
- 7
Gemini 3.1 Pro ↓4 16.0
- 8
Seed2.1 Pro ↑3 15.0
- 9
Qwen 3.8 Max ↓3 15.0
- 10
DeepSeek-V4-Pro ↓2 14.0
Seed 104
19 models we track have a score here.
- 1 Claude Opus 5.5 22.0
- 2
Gemini 3.1 Pro ↑1 19.0
- 3
Qwen 3.8 Max ↑3 18.0
- 4 GPT-6 Astra ↓2 18.0
- 5 GPT-6 Sol 17.0
- 6
DeepSeek-V4-Pro ↑2 16.0
- 7 Claude Fable 5.1 ↓3 16.0
- 8
Gemini 3.8 Flash ↓1 15.0
- 9
Tencent Hy3 13.0
- 10 Claude Sonnet 5.5 ↑3 13.0
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Hanabi Agent Benchmark leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official Hanabi Agent Benchmark 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
- 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 10 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
- Pass Independent runs Scores from the official leaderboard (21 entries).
- Pass Models scored 19 models we track, plus 2 other entries on its leaderboard.
- Pass Newest score The newest is from October 2026, for Claude Haiku 5.5.
- Pass Frontier models 4 of the top 10 have a score, led by GPT-6 Astra (#4).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on huggingface.co: 19 rows for models we track and 2 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.