Benchmark Almanac

AgenticSTS

Gemini 3.1 Pro 60% leads since June 2026
Not in model cards

Agents play the Silent in Slay the Spire 2 at its lowest difficulty. The authors compare wins with different memory settings and two other agents.

The repository has 115 stars. Its five main settings each have ten completed games. Two outside agents each played five games with Gemini 3.1 Pro.

Released
June 2026
Kind
Video games
Level
Everyday
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Actions
Interaction
Agentic
Environment
Slay the Spire 2
Items from
Generated
Refresh
One fixed set
Run by
Anyone
In views
Games, Agents
Strongest model tested
Gemini 3.1 Pro #40 on the capability index
Last model added
April 2026
Scores refreshed
November 20th, 2026
3
Models scored on its own leaderboard

Gemini 3.1 Pro leads with 60% among models released by April 2026

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart.

The newest model with a score here came out in April 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Gemini 3.1 Pro mode-a 60%
  2. 2 Qwen3.6 27B baseline-strict 0%
  3. 3 DeepSeek-V4-Pro baseline-strict 0%

Source: Official AgenticSTS leaderboard

#ModelScore ±SettingDateSource
1 Gemini 3.1 Pro 60% 31 to 83 mode-a, AgenticSTS Feb 2026 Official leaderboard
2 Qwen3.6 27B 0% baseline-strict, AgenticSTS Apr 2026 Official leaderboard
3 DeepSeek-V4-Pro 0% baseline-strict, AgenticSTS Apr 2026 Official leaderboard

Scores at each effort level

Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.

Lowest effort to highest, light to dark
Gemini 3.1 Pro AgenticSTS
baseline-st… prompt-only mode-a
60%
Qwen3.6 27B AgenticSTS
baseline-st…
0%
DeepSeek-V4-Pro AgenticSTS
baseline-st…
0%

Source: Official AgenticSTS leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Gemini 3.1 Pro mode-a AgenticSTS 60%
Gemini 3.1 Pro mode-b-frozen AgenticSTS 60%
Gemini 3.1 Pro full-frozen AgenticSTS 60%
Gemini 3.1 Pro prompt-only AgenticSTS 40%
Gemini 3.1 Pro baseline-strict AgenticSTS 30%
Gemini 3.1 Pro Silent A0 STS2MCP 0%
Gemini 3.1 Pro Silent A0 CharTyr 0%
Qwen3.6 27B baseline-strict AgenticSTS 0%
Qwen3.6 27B full-frozen AgenticSTS 0%
DeepSeek-V4-Pro baseline-strict AgenticSTS 0%
DeepSeek-V4-Pro full-frozen AgenticSTS 0%

Number one over time

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

Google DeepMind
Own leaderboard Not tested since Jun 2026

Source: Official AgenticSTS leaderboard

Scores over time

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

Plot with other benchmarks
On the benchmark’s own leaderboard
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3.1 Pro 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3.1 Pro

Source: Official AgenticSTS 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

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 7 months ago, in April 2026, for DeepSeek-V4-Pro. 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 (3 entries).
  • Note Models scored Only 3 models we track.
  • Warning Newest score The newest came 7 months ago, in April 2026, for DeepSeek-V4-Pro.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3.1 Pro (#40).
  • Pass Headroom The best is 60%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 11 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.

How we check