Benchmark Almanac

SnorkelWordle

GPT-5 94% leads since August 2025
Not in model cards

Models play Wordle on 100 rare five-letter English words, with six guesses and letter feedback after each guess. The score is the win rate.

Target words were picked with the Google Books Ngram corpus to avoid common words. The repo dates from July 30th, 2025. The board is archived.

Made by
Snorkel AI
Released
July 2025
Kind
Puzzles and chess
Level
Everyday
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Strongest model tested
Gemini 3 Pro #55 on the capability index
Last model added
November 2025
Scores refreshed
November 12th, 2026
21
Entries on its own leaderboard, 19 models we track

GPT-5 leads with 94% among models released by November 2025

Best score per entry from the benchmark's own leaderboard, read on November 12th, 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 newest model with a score here came out in November 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

21 entries. Scroll sideways for the rest.

  1. 94.0 GPT-5
  2. 93.0 Grok 4
  3. 92.9 o3
  4. 91.9 o4-mini
  5. 91.0 GPT-5 mini
  6. 91.0 Gemini 3 Pro
  7. 90.0 o3-mini
  8. 88.0 Grok 4 Fast
  9. 85.6 Claude Opus 4
  10. 85.0 Kimi K2 Thinking
  11. 83.0 Claude Sonnet 4
  12. 81.6 gpt-oss-120b
  13. 79.0 GPT-5 nano
  14. 74.0 Gemini 2.5 Pro [snapshot unspecified]
  15. 71.0 Grok 3
  16. 68.0 Claude 3.7 Sonnet
  17. 65.9 gpt-oss-20b
  18. 62.0 GPT-4.1
  19. 61.9 Gemini 2.5 Flash [snapshot unspecified]
  20. 54.0 Kimi K2
  21. 10.2 Llama 3.3 70B
  1. 1 GPT-5 94%
  2. 2 Grok 4 93%
  3. 3 o3 92.9%
  4. 4 o4-mini 91.9%
  5. 5 GPT-5 mini 91%
  6. 6 Gemini 3 Pro 91%
  7. 7 o3-mini 90%
  8. 8 Grok 4 Fast reasoning 88%
  9. 9 Claude Opus 4 85.6%
  10. 10 Kimi K2 Thinking 85%
  11. 11 Claude Sonnet 4 83%
  12. 12 gpt-oss-120b 81.6%
  13. 13 GPT-5 nano 79%
  14. 14 Gemini 2.5 Pro [snapshot unspecified] 74%
  15. 15 Grok 3 71%
  16. 16 Claude 3.7 Sonnet 68%
  17. 17 gpt-oss-20b 65.9%
  18. 18 GPT-4.1 62%
  19. 19 Gemini 2.5 Flash [snapshot unspecified] 61.9%
  20. 20 Kimi K2 Sep 2025 54%
  21. 21 Llama 3.3 70B 10.2%

Source: Official SnorkelWordle leaderboard

#EntryScoreSettingDateSource
1 GPT-5 94% Aug 2025 Official leaderboard
2 Grok 4 93% Jul 2025 Official leaderboard
3 o3 92.9% Apr 2025 Official leaderboard
4 o4-mini 91.9% Apr 2025 Official leaderboard
5 GPT-5 mini 91% Aug 2025 Official leaderboard
6 Gemini 3 Pro 91% Nov 2025 Official leaderboard
7 o3-mini 90% Jan 2025 Official leaderboard
8 Grok 4 Fast 88% reasoning Sep 2025 Official leaderboard
9 Claude Opus 4 85.6% May 2025 Official leaderboard
10 Kimi K2 Thinking 85% Nov 2025 Official leaderboard
11 Claude Sonnet 4 83% May 2025 Official leaderboard
12 gpt-oss-120b 81.6% Aug 2025 Official leaderboard
13 GPT-5 nano 79% Aug 2025 Official leaderboard
14 Gemini 2.5 Pro [snapshot unspecified] 74% Official leaderboard
15 Grok 3 71% Apr 2025 Official leaderboard
16 Claude 3.7 Sonnet 68% Feb 2025 Official leaderboard
17 gpt-oss-20b 65.9% Aug 2025 Official leaderboard
18 GPT-4.1 62% Apr 2025 Official leaderboard
19 Gemini 2.5 Flash [snapshot unspecified] 61.9% Official leaderboard
20 Kimi K2 Sep 2025 54% Sep 2025 Official leaderboard
21 Llama 3.3 70B 10.2% Dec 2024 Official leaderboard

3 models have held number one on the official leaderboard since its release in July 2025

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

OpenAI xAI
Own leaderboard Passed 85% in Jul 2025, by o3
GPT-5 94%

Source: Official SnorkelWordle leaderboard

O3-mini on the official leaderboard passed 85% in January 2025

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 Best so far
0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) GPT-5 o3 Llama 3.3 70B 0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) GPT-5 Llama 3.3 70B

Source: Official SnorkelWordle leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Games and long tasks

All 82 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 Astra 99.95%
  • 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%