CrossWordBench
Models solve generated crossword grids from text or images. The tasks vary the grid size and the share of letters already filled in.
The Markovian Thinker team independently ran CrossWordBench to test Delethink. The authors publish the generator and require registration for the hosted data.
- Made by
- HINT-lab and collaborators
- Released
- March 2025
- Kind
- Puzzles and chess
- Level
- School
- Scored
- Checked automatically
- Answers
- Short answer
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Items from
- Generated
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision
- Strongest model tested
- Claude 3.7 Sonnet #119 on the capability index
- Last model added
- March 2025
- Scores refreshed
- November 20th, 2026
DeepSeek-R1 leads with 70.7% among models released by March 2025
Best score per model from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The headline averages word correctness across the five puzzle sets. Where a model ran with several settings or harnesses, we keep its best.
The newest model with a score here came out in March 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
- 1
DeepSeek-R1 Text, zero-shot CoT 70.7%
- 2 o3-mini Text, zero-shot CoT, high reasoning effort 68%
- 3
QwQ-32B Text, zero-shot CoT 53%
- 4 Claude 3.7 Sonnet Text, zero-shot CoT, thinking off 50.6%
- 5 GPT-4o Nov 2024, Text, zero-shot CoT 46.1%
- 6
Pixtral Large Image, zero-shot CoT 29.1%
- 7
Qwen2.5-VL-72B Image, zero-shot CoT 29%
- 8
Gemini 2.0 Flash Feb 2025, Image, zero-shot CoT 26%
Source: The CrossWordBench paper
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 70.7% | Text, zero-shot CoT | Jan 2025 | Official leaderboard |
| 2 | o3-mini | 68% | Text, zero-shot CoT, high reasoning effort | Jan 2025 | Official leaderboard |
| 3 | | 53% | Text, zero-shot CoT | Mar 2025 | Official leaderboard |
| 4 | Claude 3.7 Sonnet | 50.6% | Text, zero-shot CoT, thinking off | Feb 2025 | Official leaderboard |
| 5 | GPT-4o Nov 2024 | 46.1% | Text, zero-shot CoT | Nov 2024 | Official leaderboard |
| 6 | | 29.1% | Image, zero-shot CoT | Nov 2024 | Official leaderboard |
| 7 | | 29% | Image, zero-shot CoT | Jan 2025 | Official leaderboard |
| 8 | | 26% | Image, zero-shot CoT | Feb 2025 | Official leaderboard |
Rankings by part
We read 5 parts from the benchmark’s paper, each ranking models on one part of the test, such as English 7x7, English 14x14 and Chinese. DeepSeek-R1 leads 3 of them, and o3-mini leads the other 2. Arrows show how far a model moved from its overall rank among the models we track.
English 7x7
8 models we track have a score here.
- 1
DeepSeek-R1 64.6%
- 2 o3-mini 58.7%
- 3 Claude 3.7 Sonnet ↑1 48.2%
- 4 GPT-4o Nov 2024 ↑1 41%
- 5
QwQ-32B ↓2 34.7%
- 6
Pixtral Large 29.7%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 27.7%
- 8
Qwen2.5-VL-72B ↓1 20.7%
English 14x14
8 models we track have a score here.
- 1
DeepSeek-R1 47.2%
- 2 Claude 3.7 Sonnet ↑2 44.6%
- 3 o3-mini ↓1 44.5%
- 4 GPT-4o Nov 2024 ↑1 35%
- 5
Gemini 2.0 Flash Feb 2025 ↑3 26%
- 6
QwQ-32B ↓3 25.4%
- 7
Pixtral Large ↓1 25.1%
- 8
Qwen2.5-VL-72B ↓1 19.4%
Chinese
8 models we track have a score here.
- 1
DeepSeek-R1 90.7%
- 2 o3-mini 77.4%
- 3
QwQ-32B 70.1%
- 4 GPT-4o Nov 2024 ↑1 59.3%
- 5 Claude 3.7 Sonnet ↓1 47.8%
- 6
Qwen2.5-VL-72B ↑1 39.1%
- 7
Pixtral Large ↓1 25.2%
- 8
Gemini 2.0 Flash Feb 2025 20.8%
English Simple
8 models we track have a score here.
- 1 o3-mini ↑1 78.2%
- 2
DeepSeek-R1 ↓1 75.9%
- 3
QwQ-32B 64.7%
- 4 Claude 3.7 Sonnet 53.9%
- 5 GPT-4o Nov 2024 43.8%
- 6
Qwen2.5-VL-72B ↑1 23.9%
- 7
Gemini 2.0 Flash Feb 2025 ↑1 22.9%
- 8
Pixtral Large ↓2 21.6%
CommonSenseQA
8 models we track have a score here.
- 1 o3-mini ↑1 81.2%
- 2
DeepSeek-R1 ↓1 75.2%
- 3
QwQ-32B 69.9%
- 4 Claude 3.7 Sonnet 58.3%
- 5 GPT-4o Nov 2024 52.4%
- 6
Pixtral Large 43.9%
- 7
Qwen2.5-VL-72B 41.8%
- 8
Gemini 2.0 Flash Feb 2025 32.7%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The CrossWordBench paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official CrossWordBench 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
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 20 months ago, in March 2025, for QwQ-32B. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the benchmark’s paper (8 entries).
- Pass Models scored 8 models we track.
- Warning Newest score The newest came 20 months ago, in March 2025, for QwQ-32B.
- Note Frontier models None of the top 10 has a score. The strongest tested is Claude 3.7 Sonnet (#119).
- Pass Headroom The best is 70.7%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 10 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.