Benchmark Almanac

Baba Is AI

Llama 3.2 90B 43.9% leads since September 2024
Not in model cards

Models see a puzzle and write a plan to reach the winning object. Some puzzles require them to push words together to change the rules.

The original authors scored exact plan matches. The BALROG team later measured progress on 40 levels with seven language models and six vision models.

Released
July 2024
Kind
Puzzles and chess
Level
Everyday
Scored
Checked automatically
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
English
In and out
Images, Text → Text
Interaction
Single turn
Environment
Baba Is You
Items from
Written by experts
Refresh
One fixed set
Run by
Anyone
In views
Games, Vision
Strongest model tested
Gemini 1.5 Pro Sep 2024 #157 on the capability index
Last model added
March 2025
Scores refreshed
November 20th, 2026
7
Models scored on BALROG’s leaderboard

Llama 3.2 90B leads with 43.9% among models released by March 2025

Best score per model from BALROG’s 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 March 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Llama 3.2 90B LLM 43.9%
  2. 2 Llama 3.1-70B LLM 40%
  3. 3 GPT-4o Mar 2025, LLM 33.7%
  4. 4 Gemini 1.5 Pro Sep 2024, LLM 32%
  5. 5 GPT-4o mini VLM 16.4%
  6. 6 Llama 3.2 11B LLM 15.6%
  7. 7 Gemini 1.5 Flash Sep 2024, LLM 12.8%

Source: BALROG’s leaderboard

#ModelScoreSettingDateSource
1 Llama 3.2 90B 43.9% LLM Sep 2024 Official leaderboard
2 Llama 3.1-70B 40% LLM Jul 2024 Official leaderboard
3 GPT-4o Mar 2025 33.7% LLM Mar 2025 Official leaderboard
4 Gemini 1.5 Pro Sep 2024 32% LLM Sep 2024 Official leaderboard
5 GPT-4o mini 16.4% VLM Jul 2024 Official leaderboard
6 Llama 3.2 11B 15.6% LLM Sep 2024 Official leaderboard
7 Gemini 1.5 Flash Sep 2024 12.8% LLM Aug 2024 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
Llama 3.2 90B
VLM LLM
43.9%
GPT-4o
VLM LLM
33.7%
Gemini 1.5 Pro
VLM
32%
GPT-4o mini
LLM
16.4%
Llama 3.2 11B
VLM LLM
15.6%
Gemini 1.5 Flash
VLM LLM
12.8%

Source: BALROG’s leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Llama 3.2 90B LLM 43.9%
Llama 3.1-70B LLM 40%
GPT-4o Mar 2025 LLM 33.7%
Gemini 1.5 Pro Sep 2024 LLM 32%
Gemini 1.5 Pro Sep 2024 VLM 31.4%
Llama 3.2 90B VLM 21.9%
GPT-4o Mar 2025 VLM 18.6%
GPT-4o mini VLM 16.4%
GPT-4o mini LLM 15.6%
Llama 3.2 11B LLM 15.6%
Gemini 1.5 Flash Sep 2024 LLM 12.8%
Gemini 1.5 Flash Sep 2024 VLM 8.30%
Llama 3.2 11B VLM 5.76%

3 models have held number one on BALROG’s leaderboard since its release in July 2024

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

OpenAI Meta
BALROG leaderboard Not tested since Mar 2025

Source: BALROG’s 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 BALROG’s leaderboard Best so far
0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Llama 3.2 90B Llama 3.1-70B GPT-4o mini 0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Llama 3.2 90B GPT-4o mini

Source: BALROG’s 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 20 months ago, in March 2025, for GPT-4o. 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 (7 entries).
  • Pass Models scored 7 models we track.
  • Warning Newest score The newest came 20 months ago, in March 2025, for GPT-4o.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 1.5 Pro (#157).
  • Pass Headroom The best is 43.9%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on BALROG: 13 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