Benchmark Almanac

ChessBench

Stockfish 16 (1.5s per board) 100% tops its leaderboard GPT-3.5 Turbo Instruct 66.5% leads the models we track
Not in model cards

Models select chess moves from legal options in positions scored by a chess engine. Decision Index measures whether the selected move has maximal action value.

The Decision Index adaptation samples 5,000 positions from searchless_chess and accepts tied best moves. This accuracy is not chess Elo. The adaptation release month remains unconfirmed.

Made by
Google DeepMind
Kind
Puzzles and chess
Level
Competition
Scored
Checked automatically
Answers
Multiple choice
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Items from
Another benchmark
Refresh
One fixed set
Run by
Anyone
Last model added
September 2023
Scores refreshed
November 21st, 2026
11
Entries in its paper, 1 model we track

Stockfish 16 (1.5s per board) tops the leaderboard with 100% GPT-3.5 Turbo Instruct leads with 66.5%

Best score per entry from the benchmark's paper, read on November 21st, 2026. A paper’s table doesn’t add newer models. 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 best model we track is GPT-3.5 Turbo Instruct, at 66.5%. This view shows only models we track, each at its best.

  1. 1 Stockfish 16 (1.5s per board) SL 100%
  2. 2 Leela Chess Zero (400 MCTS sim.) RL 99.6%
  3. 3 Leela Chess Zero (value net only) RL 95.9%
  4. 4 AlphaZero (400 MCTS sim.) RL 95.6%
  5. 5 270M Transformer SL 95.4%
  6. 6 136M Transformer SL 94.5%
  7. 7 9M Transformer SL 88.9%
  8. 8 Leela Chess Zero (policy net only) RL 88.6%
  9. 9 AlphaZero (value net only) RL 82%
  10. 10 GPT-3.5 Turbo Instruct SSL 66.5%
  11. 11 AlphaZero (policy net only) RL 56.1%

Source: The ChessBench paper

Source: The ChessBench paper

#EntryScoreSettingDateSource
1 Stockfish 16 (1.5s per board) 100% SL Official leaderboard
2 Leela Chess Zero (400 MCTS sim.) 99.6% RL Official leaderboard
3 Leela Chess Zero (value net only) 95.9% RL Official leaderboard
4 AlphaZero (400 MCTS sim.) 95.6% RL Official leaderboard
5 270M Transformer 95.4% SL Official leaderboard
6 136M Transformer 94.5% SL Official leaderboard
7 9M Transformer 88.9% SL Official leaderboard
8 Leela Chess Zero (policy net only) 88.6% RL Official leaderboard
9 AlphaZero (value net only) 82% RL Official leaderboard
10 GPT-3.5 Turbo Instruct 66.5% SSL Sep 2023 Official leaderboard
11 AlphaZero (policy net only) 56.1% RL Official leaderboard
#ModelScoreSettingDateSource
1 GPT-3.5 Turbo Instruct 66.5% SSL Sep 2023 Official leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

In the benchmark’s paper
0% 20% 40% 60% 80% 100% Near the ceiling (85%) GPT-3.5 Turbo Instruct 0% 20% 40% 60% 80% 100% Near the ceiling (85%) GPT-3.5 Turbo Instruct

Source: Official ChessBench 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

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 3 years ago, in September 2023, for GPT-3.5 Turbo Instruct. 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 (11 entries).
  • Pass Models scored 1 model we track, plus 10 other entries on its leaderboard.
  • Warning Newest score The newest came 3 years ago, in September 2023, for GPT-3.5 Turbo Instruct.
  • Note Frontier models None of its models are on the capabilities index.
  • Pass Headroom The best is 66.5%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 1 row for models we track and 10 other entries at the last read.
  • Pass Last read November 21st, 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