Benchmark Almanac

BuilderBench

GPT-6 Astra 98% leads since September 2026
Not in model cards

Agents learn to control a simulated robot that builds target block structures through physical interaction and exploration.

The task suite contains more than 50 block-building tasks and a public leaderboard.

Released
October 2025
Kind
Embodied reasoning
Level
Expert
Scored
Settled by events
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text, 3D → Actions
Interaction
Agentic
Environment
A simulator
Items from
Generated
Refresh
One fixed set
Run by
Anyone
In views
Agents, Robotics
Strongest model tested
GPT-6 Astra #4 on the capability index
Last model added
September 2026
Scores refreshed
November 21st, 2026
5
Models scored on its own leaderboard

GPT-6 Astra leads with 98% among models released by September 2026

Best score per model from the benchmark's own leaderboard, read on November 21st, 2026. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in September 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-6 Astra 98%
  2. 2 Claude Opus 4.8 91.8%
  3. 3 Gemini 3 Flash 47.1%
  4. 4 Claude Opus 4.6 47.1%
  5. 5 GPT-5.2 29.4%

Source: Official BuilderBench leaderboard

#ModelScoreSettingReleasedRun onSource
1 GPT-6 Astra 98% codex Sep 2026 Sep 2026 Official leaderboard
2 Claude Opus 4.8 91.8% claude code May 2026 Sep 2026 Official leaderboard
3 Gemini 3 Flash 47.1% reflexion Dec 2025 Mar 2026 Official leaderboard
4 Claude Opus 4.6 47.1% reflexion Feb 2026 Mar 2026 Official leaderboard
5 GPT-5.2 29.4% cot Dec 2025 Mar 2026 Official leaderboard

Rankings by part

We read 2 parts from the benchmark’s own leaderboard: Easy tasks and Hard tasks. Each ranks models on one part of the test. Claude Opus 4.8 leads Easy tasks and GPT-6 Astra leads Hard tasks. Arrows show how far a model moved from its overall rank among the models we track.

Easy tasks

5 models we track have a score here.

  1. 1 Claude Opus 4.8 ↑1 100%
  2. 2 GPT-6 Astra ↓1 100%
  3. 3 Gemini 3 Flash 66.7%
  4. 4 Claude Opus 4.6 62.5%
  5. 5 GPT-5.2 45.8%

4 models have held number one on the official leaderboard since its release in October 2025

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

OpenAI Google DeepMind Anthropic
Own leaderboard Passed 85% in May 2026, by Claude Opus 4.8

Source: Official BuilderBench leaderboard

Claude Opus 4.8 on the official leaderboard passed 85% in May 2026

That was 7 months after its release. By the same age, 15% of the benchmarks from 2025 had passed 85%. 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% 2026 Near the ceiling (85%) GPT-6 Astra Claude Opus 4.8 Gemini 3 Flash GPT-5.2 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) GPT-6 Astra Gemini 3 Flash

Source: Official BuilderBench 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 Robotics

All 172 in Robotics
  • ERQA In use 400 multiple-choice visual questions from robot and egocentric footage on spatial, trajectory, action and state reasoning. Accuracy. Gemini Robotics ER 2 78.5%
  • SimplerEnv In use Real-to-sim replicas of Google Robot and WidowX/Bridge setups. Policies scored by success rate on pick, move, drawer and Bridge tasks. Praxis-1 80%
  • RoboLab In use Zero-shot DROID Franka policies in Isaac Lab on 120 tabletop tasks at three instruction levels, 10 episodes each. Success rate. Praxis-1 46.2%
  • RoboDojo In use Generalist manipulation policies run 42 Isaac Sim tasks and 18 remote real-robot tasks. Scored by task score and success rate. GPT-6 Astra 27.6%
  • Progress classification (Google DeepMind) In use Models assign each frame of a robot video to one of five task-progress bands from 0 to 100 percent. Scored by accuracy. Gemini Robotics ER 2 57.4%
  • Human proximity (Google DeepMind robotics) In use Models judge from robot camera views whether a human is within 1 meter, a spatial awareness check for safe robot behavior. Scored by accuracy. Gemini Robotics ER 2 93%

Data audit

All 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (5 entries).
  • Pass Models scored 5 models we track.
  • Pass Newest score The newest is from September 2026, for GPT-6 Astra.
  • Note Frontier models 1 of the top 10 has a score, which is fine for a retired test. The strongest tested is GPT-6 Astra (#4).
  • Note Headroom Saturated. The best is 98%, past 85% since May 2026.
Leaderboard import 3 pass
  • Pass Importer We read the board on rajghugare19.github.io: 5 rows for models we track 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