ScrabbleBench
Models choose a word from a Scrabble rack and board. A solver finds its best legal placement. The score averages the four seats in self-play.
The maker ranks 17 models on one fixed seed. Four finalists also played a separate tournament with rotating seats.
- Made by
- Sunny Balasubramanian
- Released
- August 2025
- Announced on X
- 57K views, 209 likes
- Kind
- Puzzles and chess
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Multi-agent
- Environment
- Scrabble
- Items from
- Generated
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
- GPT-5 #71 on the capability index
- Last model added
- September 2025
- Scores refreshed
- November 20th, 2026
GPT-5 leads with 152.8 among models released by September 2025 GPT-5 leads with 152.8 among models released by September 2025
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.
The newest model with a score here came out in September 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 152.8 GPT-5
- 145.8
Gemini 2.5 Pro
- 122.0 Claude Sonnet 4
- 121.2 Grok 3
- 121.0
Qwen3-235B-A22B-Thinking
- 116.2 Grok 4
- 115.8 GPT-4o
- 114.8 Claude Opus 4.1
- 112.8
Gemini 2.5 Flash
- 109.8 google/gemini-2.0-flash-001
- 105.5 google/gemini-2.0-flash-lite-001
- 104.0 deepseek/deepseek-chat-v3-0324
- 102.0 Kimi K2
- 101.0 GPT-4.1
- 89.8 gpt-oss-120b
- 85.0
Qwen3-30B-A3B-Instruct
- 65.0 GPT-5 nano
- 1 GPT-5 152.8
- 2
Gemini 2.5 Pro Jun 2025 145.8
- 3 Claude Sonnet 4 122.0
- 4 Grok 3 121.2
- 5
Qwen3-235B-A22B-Thinking Jul 2025 121.0
- 6 Grok 4 116.2
- 7 GPT-4o Mar 2025 115.8
- 8 Claude Opus 4.1 114.8
- 9
Gemini 2.5 Flash Sep 2025 112.8
- 10 google/gemini-2.0-flash-001 109.8
- 11 google/gemini-2.0-flash-lite-001 105.5
- 12 deepseek/deepseek-chat-v3-0324 104.0
- 13 Kimi K2 Sep 2025 102.0
- 14 GPT-4.1 101.0
- 15 gpt-oss-120b 89.8
- 16
Qwen3-30B-A3B-Instruct Jul 2025 85.0
- 17 GPT-5 nano 65.0
Source: Official ScrabbleBench leaderboard
- 1 GPT-5 152.8
- 2
Gemini 2.5 Pro Jun 2025 145.8
- 3 Claude Sonnet 4 122.0
- 4 Grok 3 121.2
- 5
Qwen3-235B-A22B-Thinking Jul 2025 121.0
- 6 Grok 4 116.2
- 7 GPT-4o Mar 2025 115.8
- 8 Claude Opus 4.1 114.8
- 9
Gemini 2.5 Flash Sep 2025 112.8
- 10 Kimi K2 Sep 2025 102.0
- 11 GPT-4.1 101.0
- 12 gpt-oss-120b 89.8
- 13
Qwen3-30B-A3B-Instruct Jul 2025 85.0
- 14 GPT-5 nano 65.0
Source: Official ScrabbleBench leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5 | 152.8 | Aug 2025 | Official leaderboard | |
| 2 | | 145.8 | Jun 2025 | Official leaderboard | |
| 3 | Claude Sonnet 4 | 122.0 | May 2025 | Official leaderboard | |
| 4 | Grok 3 | 121.2 | Apr 2025 | Official leaderboard | |
| 5 | | 121.0 | Jul 2025 | Official leaderboard | |
| 6 | Grok 4 | 116.2 | Jul 2025 | Official leaderboard | |
| 7 | GPT-4o Mar 2025 | 115.8 | Mar 2025 | Official leaderboard | |
| 8 | Claude Opus 4.1 | 114.8 | Aug 2025 | Official leaderboard | |
| 9 | | 112.8 | Sep 2025 | Official leaderboard | |
| 10 | google/gemini-2.0-flash-001 | 109.8 | Official leaderboard | ||
| 11 | google/gemini-2.0-flash-lite-001 | 105.5 | Official leaderboard | ||
| 12 | deepseek/deepseek-chat-v3-0324 | 104.0 | Official leaderboard | ||
| 13 | Kimi K2 Sep 2025 | 102.0 | Sep 2025 | Official leaderboard | |
| 14 | GPT-4.1 | 101.0 | Apr 2025 | Official leaderboard | |
| 15 | gpt-oss-120b | 89.8 | Aug 2025 | Official leaderboard | |
| 16 | | 85.0 | Jul 2025 | Official leaderboard | |
| 17 | GPT-5 nano | 65.0 | Aug 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5 | 152.8 | Aug 2025 | Official leaderboard | |
| 2 | | 145.8 | Jun 2025 | Official leaderboard | |
| 3 | Claude Sonnet 4 | 122.0 | May 2025 | Official leaderboard | |
| 4 | Grok 3 | 121.2 | Apr 2025 | Official leaderboard | |
| 5 | | 121.0 | Jul 2025 | Official leaderboard | |
| 6 | Grok 4 | 116.2 | Jul 2025 | Official leaderboard | |
| 7 | GPT-4o Mar 2025 | 115.8 | Mar 2025 | Official leaderboard | |
| 8 | Claude Opus 4.1 | 114.8 | Aug 2025 | Official leaderboard | |
| 9 | | 112.8 | Sep 2025 | Official leaderboard | |
| 10 | Kimi K2 Sep 2025 | 102.0 | Sep 2025 | Official leaderboard | |
| 11 | GPT-4.1 | 101.0 | Apr 2025 | Official leaderboard | |
| 12 | gpt-oss-120b | 89.8 | Aug 2025 | Official leaderboard | |
| 13 | | 85.0 | Jul 2025 | Official leaderboard | |
| 14 | GPT-5 nano | 65.0 | Aug 2025 | Official leaderboard |
Rankings by part
We read 3 parts from the benchmark’s own leaderboard: Easy, Medium and Hard. Each ranks models on one part of the test. GPT-4o leads 1 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
Easy
14 models we track have a score here.
- 1 GPT-4o Mar 2025 ↑6 100%
- 2 Claude Sonnet 4 ↑1 100%
- 3
Gemini 2.5 Pro Jun 2025 ↓1 100%
- 4 GPT-5 ↓3 100%
- 5 Claude Opus 4.1 ↑3 93%
- 6 Grok 4 92%
- 7
Gemini 2.5 Flash Sep 2025 ↑2 89%
- 8
Qwen3-235B-A22B-Thinking Jul 2025 ↓3 79%
- 9 GPT-4.1 ↑2 72%
- 10
Qwen3-30B-A3B-Instruct Jul 2025 ↑3 67%
Medium
14 models we track have a score here.
- 1 Claude Sonnet 4 ↑2 100%
- 2 GPT-5 ↓1 100%
- 3
Gemini 2.5 Pro Jun 2025 ↓1 88%
- 4 GPT-4o Mar 2025 ↑3 86%
- 5 Claude Opus 4.1 ↑3 80%
- 6 Grok 4 79%
- 7
Qwen3-30B-A3B-Instruct Jul 2025 ↑6 69%
- 8 GPT-4.1 ↑3 60%
- 9
Gemini 2.5 Flash Sep 2025 60%
- 10
Qwen3-235B-A22B-Thinking Jul 2025 ↓5 54%
Hard
14 models we track have a score here.
- 1 GPT-5 92%
- 2
Gemini 2.5 Pro Jun 2025 88%
- 3 Claude Sonnet 4 67%
- 4 Claude Opus 4.1 ↑4 67%
- 5 Grok 4 ↑1 60%
- 6
Qwen3-235B-A22B-Thinking Jul 2025 ↓1 57%
- 7
Qwen3-30B-A3B-Instruct Jul 2025 ↑6 50%
- 8 GPT-4o Mar 2025 ↓1 42%
- 9 gpt-oss-120b ↑3 40%
- 10 Grok 3 ↓6 39%
2 models have held number one on the official leaderboard since its release in August 2025
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official ScrabbleBench 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 ScrabbleBench 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
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 14 months ago, in September 2025, for Gemini 2.5 Flash. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (17 entries).
- Pass Models scored 14 models we track, plus 3 other entries on its leaderboard.
- Warning Newest score The newest came 14 months ago, in September 2025, for Gemini 2.5 Flash.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5 (#71).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on sunnybala.github.io: 14 rows for models we track and 3 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.