Benchmark Almanac

SpireClimb

GPT-6 Astra 34% leads since October 2026
Not in model cards

Agents play Slay the Spire 1 and 2 at max difficulty through a computer-use harness on fixed seeds. Score is the mean win rate.

Top human players win ~85% (StS1) and ~95% (StS2). Seeds are not published.

Made by
Epoch AI
Released
October 2026
Kind
Video games
Level
Expert
Scored
Checked automatically
Answers
Agent task
Tasks
Kept private
Strongest model tested
Claude Fable 5.2 #1 on the capability index
Last model added
November 2026
Scores refreshed
November 12th, 2026
9
Models scored in Epoch AI’s runs

GPT-6 Astra leads with 34%

Best score per model from Epoch AI’s own runs. Where a model ran with several settings or harnesses, we keep its best.

  1. 34.0 GPT-6 Astra
  2. 23.8 GPT-6.1 Sol
  3. 22.3 GPT-6.1 Terra
  4. 19.6 Claude Fable 5.2
  5. 9.17 GPT-6 Sol
  6. 5.83 Claude Opus 5.5
  7. 2.00 Claude Fable 5.1
  8. 1.33 Claude Sonnet 5.5
  9. 0.0 GPT-6 Luna
  1. 1 GPT-6 Astra max 34%
  2. 2 GPT-6.1 Sol max 23.8%
  3. 3 GPT-6.1 Terra max 22.3%
  4. 4 Claude Fable 5.2 max 19.6%
  5. 5 GPT-6 Sol max 9.17%
  6. 6 Claude Opus 5.5 max 5.83%
  7. 7 Claude Fable 5.1 max 2.00%
  8. 8 Claude Sonnet 5.5 max 1.33%
  9. 9 GPT-6 Luna max 0%

Source: Epoch AI runs (CC BY 4.0)

#ModelScoreSettingDateSource
1 GPT-6 Astra 34% max Sep 2026 Epoch AI run
2 GPT-6.1 Sol 23.8% max Sep 2026 Epoch AI run
3 GPT-6.1 Terra 22.3% max Oct 2026 Epoch AI run
4 Claude Fable 5.2 19.6% max Nov 2026 Epoch AI run
5 GPT-6 Sol 9.17% max Sep 2026 Epoch AI run
6 Claude Opus 5.5 5.83% max Sep 2026 Epoch AI run
7 Claude Fable 5.1 2.00% max Sep 2026 Epoch AI run
8 Claude Sonnet 5.5 1.33% max Sep 2026 Epoch AI run
9 GPT-6 Luna 0% max Sep 2026 Epoch AI run

Number one over time

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

OpenAI
Epoch AI runs 1 leader
GPT-6 Astra 34%

Source: Epoch AI runs (CC BY 4.0)

Scores over time

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

Run by Epoch AI Best so far
0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) GPT-6 Astra Claude Fable 5.1 0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) GPT-6 Astra Claude Fable 5.1

Source: Epoch AI runs (CC BY 4.0)

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%