Benchmark Almanac

ParliamentBench

GPT-5.4 81% leads since July 2026
Not in model cards

Models play Secret Hitler while deceiving, persuading, and reasoning under hidden roles.

The July 2026 paper evaluates 16 language models across 1,600 simulated matches. The abstract does not state the best score.

Released
July 2026
Kind
Multi-agent games
Level
Everyday
Scored
Settled by events
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text, Actions
Interaction
Multi-agent
Environment
Secret Hitler
Refresh
Live
Run by
Anyone
In views
Games, Agents, Kept fresh
Strongest model tested
GPT-5.4 #28 on the capability index
Last model added
March 2026
Scores refreshed
November 20th, 2026
12
Models scored on its own leaderboard

GPT-5.4 leads with 81% among models released by March 2026

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. The score is the percentage of games the model won against Llama 3.3 70B opponents. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in March 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 GPT-5.4 81%
  2. 2 Kimi K2.5 76%
  3. 3 Grok 4.1 Fast 69%
  4. 4 DeepSeek-V3.1-Terminus 66%
  5. 5 Llama 3.3 70B 57.6%
  6. 6 Qwen3.5 397B-A17B 56%
  7. 7 Olmo 3.1 32B Instruct 53%
  8. 8 gpt-oss-120b 50%
  9. 9 Mistral Small 3 43%
  10. 10 Llama 3.1-70B 42%
  11. 11 Gemma 3 27B 39%
  12. 12 gpt-oss-20b 24%

Source: Official ParliamentBench leaderboard

#ModelScore ±SettingDateSource
1 GPT-5.4 81% 73 to 88 Mar 2026 Official leaderboard
2 Kimi K2.5 76% 67 to 84 Jan 2026 Official leaderboard
3 Grok 4.1 Fast 69% ±9 Nov 2025 Official leaderboard
4 DeepSeek-V3.1-Terminus 66% ±9 Sep 2025 Official leaderboard
5 Llama 3.3 70B 57.6% 48.5 to 65.7 Dec 2024 Official leaderboard
6 Qwen3.5 397B-A17B 56% ±10 Feb 2026 Official leaderboard
7 Olmo 3.1 32B Instruct 53% 44 to 63 Dec 2025 Official leaderboard
8 gpt-oss-120b 50% ±10 Aug 2025 Official leaderboard
9 Mistral Small 3 43% ±10 Jan 2025 Official leaderboard
10 Llama 3.1-70B 42% ±9 Jul 2024 Official leaderboard
11 Gemma 3 27B 39% ±9 Mar 2025 Official leaderboard
12 gpt-oss-20b 24% 16 to 33 Aug 2025 Official leaderboard

Rankings by part

We read 3 parts from the benchmark’s own leaderboard: Liberal, Fascist and Hitler. Each ranks models on one part of the test. GPT-5.4 leads 2 of them, and Kimi K2.5 leads the other one. Arrows show how far a model moved from its overall rank among the models we track.

Liberal

12 models we track have a score here.

  1. 1 GPT-5.4 81.7%
  2. 2 DeepSeek-V3.1-Terminus ↑2 76.7%
  3. 3 Grok 4.1 Fast 75%
  4. 4 Llama 3.3 70B ↑1 74.6%
  5. 5 Kimi K2.5 ↓3 73.3%
  6. 6 Qwen3.5 397B-A17B 60%
  7. 7 Llama 3.1-70B ↑3 55%
  8. 8 Olmo 3.1 32B Instruct ↓1 55%
  9. 9 Mistral Small 3 45%
  10. 10 Gemma 3 27B ↑1 45%

Number one over time

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

OpenAI
Own leaderboard Not tested since Jul 2026
GPT-5.4 81%

Source: Official ParliamentBench 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 Best so far
0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) GPT-5.4 Kimi K2.5 Grok 4.1 Fast Llama 3.3 70B Llama 3.1-70B 0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) GPT-5.4 Grok 4.1 Fast

Source: Official ParliamentBench 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 8 months ago, in March 2026, for GPT-5.4. 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 official leaderboard (12 entries).
  • Pass Models scored 12 models we track.
  • Warning Newest score The newest came 8 months ago, in March 2026, for GPT-5.4.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.4 (#28).
  • Pass Headroom The best is 81%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 12 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