RedstoneBench
Agents design redstone circuits in Minecraft. A game server tests whether each circuit works and measures the space it occupies.
Adam Holter maintains a public leaderboard with replayable builds for 10 models, including GPT-6.2 Astra and Gemini 4 Neon.
- Made by
- Adam Holter
- Released
- November 2026
- Kind
- Video games
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- Minecraft
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
- GPT-6.2 Astra #1 on the capability index
- Last model added
- November 2026
- Scores refreshed
- November 20th, 2026
GPT-6.2 Astra leads with 88.7
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.
- 1 GPT-6.2 Astra max 88.7
- 2 GPT-6.2 Sol max 75.8
- 3 GPT-6 Astra ultra 63.9
- 4
Gemini 4 Neon high 63.4
- 5 GPT-6.1 Sol ultra 62.5
- 6
Gemini 4 Carbon high 59.4
- 7 Claude Opus 5.5 medium 57.1
- 8 GPT-6.1 Terra max 49.5
- 9 Grok 4.7 high 29.2
- 10 GPT-6 Luna max 19.2
Source: Official RedstoneBench leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-6.2 Astra | 88.7 | max | Nov 2026 | Official leaderboard |
| 2 | GPT-6.2 Sol | 75.8 | max | Nov 2026 | Official leaderboard |
| 3 | GPT-6 Astra | 63.9 | ultra | Sep 2026 | Official leaderboard |
| 4 | | 63.4 | high | Nov 2026 | Official leaderboard |
| 5 | GPT-6.1 Sol | 62.5 | ultra | Sep 2026 | Official leaderboard |
| 6 | | 59.4 | high | Oct 2026 | Official leaderboard |
| 7 | Claude Opus 5.5 | 57.1 | medium | Sep 2026 | Official leaderboard |
| 8 | GPT-6.1 Terra | 49.5 | max | Oct 2026 | Official leaderboard |
| 9 | Grok 4.7 | 29.2 | high | Sep 2026 | Official leaderboard |
| 10 | GPT-6 Luna | 19.2 | max | Sep 2026 | Official leaderboard |
Rankings by part
We read 16 parts from the benchmark’s own leaderboard in 5 groups: doors, piston extenders, logic, memory and timing. Each ranks models on one part of the test. GPT-6.2 Astra leads 10 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
2x2 piston door
10 models we track have a score here.
- 1 GPT-6.1 Terra ↑7 100.0
- 2
Gemini 4 Carbon ↑4 100.0
- 3 GPT-6.2 Astra ↓2 100.0
- 4 GPT-6.2 Sol ↓2 100.0
- 5
Gemini 4 Neon ↓1 100.0
- 6 GPT-6 Astra ↓3 88.9
- 7 Claude Opus 5.5 88.9
- 8 GPT-6.1 Sol ↓3 88.9
- 9 Grok 4.7 66.7
- 10 GPT-6 Luna 66.7
3x3 piston door
10 models we track have a score here.
- 1 GPT-6.2 Astra 63.6
- 2 GPT-6.2 Sol 30.8
- 3 GPT-6 Astra 7.78
- 4 GPT-6.1 Sol ↑1 5.59
- 5
Gemini 4 Neon ↓1 2.56
- 6 Grok 4.7 ↑3 0.00
- 7 Claude Opus 5.5 0.00
- 8 GPT-6 Luna ↑2 0.00
- 9 GPT-6.1 Terra ↓1 0.00
- 10
Gemini 4 Carbon ↓4 0.00
4x4 piston door
10 models we track have a score here.
- 1 GPT-6.2 Astra 55.0
- 2 GPT-6.2 Sol 11.5
- 3 GPT-6 Astra 7.86
- 4 GPT-6.1 Sol ↑1 7.22
- 5
Gemini 4 Neon ↓1 3.68
- 6 Grok 4.7 ↑3 0.00
- 7 Claude Opus 5.5 0.00
- 8 GPT-6 Luna ↑2 0.00
- 9 GPT-6.1 Terra ↓1 0.00
- 10
Gemini 4 Carbon ↓4 0.00
Double piston extender
10 models we track have a score here.
- 1 GPT-6.2 Astra 100.0
- 2
Gemini 4 Neon ↑2 100.0
- 3 Claude Opus 5.5 ↑4 80.0
- 4
Gemini 4 Carbon ↑2 80.0
- 5 GPT-6 Astra ↓2 66.7
- 6 GPT-6.1 Sol ↓1 66.7
- 7 GPT-6.2 Sol ↓5 66.7
- 8 GPT-6.1 Terra 40.0
- 9 Grok 4.7 0.00
- 10 GPT-6 Luna 0.00
Triple piston extender
10 models we track have a score here.
- 1 GPT-6.2 Astra 100.0
- 2 GPT-6.2 Sol 92.3
- 3 Claude Opus 5.5 ↑4 33.3
- 4 GPT-6.1 Sol ↑1 33.3
- 5
Gemini 4 Neon ↓1 31.2
- 6
Gemini 4 Carbon 14.2
- 7 GPT-6 Astra ↓4 10.7
- 8 GPT-6.1 Terra 2.08
- 9 Grok 4.7 0.00
- 10 GPT-6 Luna 0.00
NOT gate
10 models we track have a score here.
- 1 GPT-6 Astra ↑2 100.0
- 2 Grok 4.7 ↑7 100.0
- 3 Claude Opus 5.5 ↑4 100.0
- 4 GPT-6 Luna ↑6 100.0
- 5 GPT-6.1 Sol 100.0
- 6 GPT-6.1 Terra ↑2 100.0
- 7
Gemini 4 Carbon ↓1 100.0
- 8 GPT-6.2 Astra ↓7 100.0
- 9 GPT-6.2 Sol ↓7 100.0
- 10
Gemini 4 Neon ↓6 100.0
XOR gate
10 models we track have a score here.
- 1 GPT-6.2 Astra 100.0
- 2 GPT-6.2 Sol 100.0
- 3
Gemini 4 Neon ↑1 100.0
- 4 GPT-6 Astra ↓1 75.0
- 5 Claude Opus 5.5 ↑2 75.0
- 6 GPT-6.1 Sol ↓1 75.0
- 7
Gemini 4 Carbon ↓1 75.0
- 8 GPT-6.1 Terra 50.0
- 9 GPT-6 Luna ↑1 10.7
- 10 Grok 4.7 ↓1 4.59
Full adder
10 models we track have a score here.
- 1 GPT-6.2 Astra 100.0
- 2 GPT-6 Astra ↑1 76.2
- 3 GPT-6.1 Sol ↑2 68.6
- 4 GPT-6.2 Sol ↓2 68.6
- 5
Gemini 4 Neon ↓1 60.0
- 6 Claude Opus 5.5 ↑1 44.4
- 7
Gemini 4 Carbon ↓1 43.6
- 8 GPT-6.1 Terra 38.1
- 9 Grok 4.7 0.00
- 10 GPT-6 Luna 0.00
2-bit adder
10 models we track have a score here.
- 1 GPT-6.2 Astra 100.0
- 2 GPT-6.1 Sol ↑3 87.5
- 3 GPT-6.2 Sol ↓1 58.3
- 4 GPT-6 Astra ↓1 55.6
- 5
Gemini 4 Neon ↓1 50.0
- 6 Claude Opus 5.5 ↑1 38.5
- 7 GPT-6.1 Terra ↑1 3.88
- 8
Gemini 4 Carbon ↓2 3.20
- 9 Grok 4.7 0.00
- 10 GPT-6 Luna 0.00
Hex to seven-segment decoder
10 models we track have a score here.
- 1 GPT-6.2 Astra 100.0
- 2 GPT-6 Astra ↑1 44.3
- 3 GPT-6.2 Sol ↓1 41.0
- 4 GPT-6.1 Sol ↑1 30.1
- 5 Claude Opus 5.5 ↑2 25.9
- 6
Gemini 4 Neon ↓2 19.4
- 7 GPT-6.1 Terra ↑1 16.4
- 8
Gemini 4 Carbon ↓2 8.10
- 9 Grok 4.7 0.00
- 10 GPT-6 Luna 0.00
T flip-flop
10 models we track have a score here.
- 1 GPT-6 Astra ↑2 100.0
- 2 Grok 4.7 ↑7 100.0
- 3 Claude Opus 5.5 ↑4 100.0
- 4 GPT-6.1 Sol ↑1 100.0
- 5 GPT-6.1 Terra ↑3 100.0
- 6
Gemini 4 Carbon 100.0
- 7 GPT-6.2 Astra ↓6 100.0
- 8 GPT-6.2 Sol ↓6 100.0
- 9
Gemini 4 Neon ↓5 100.0
- 10 GPT-6 Luna 0.00
Set-reset latch
10 models we track have a score here.
- 1 GPT-6 Astra ↑2 100.0
- 2 Claude Opus 5.5 ↑5 100.0
- 3 GPT-6.2 Astra ↓2 100.0
- 4 GPT-6.2 Sol ↓2 100.0
- 5
Gemini 4 Neon ↓1 100.0
- 6 GPT-6.1 Sol ↓1 87.5
- 7 GPT-6.1 Terra ↑1 70.0
- 8
Gemini 4 Carbon ↓2 70.0
- 9 GPT-6 Luna ↑1 43.8
- 10 Grok 4.7 ↓1 38.9
2-bit counter
10 models we track have a score here.
- 1 GPT-6 Astra ↑2 100.0
- 2 Claude Opus 5.5 ↑5 100.0
- 3 GPT-6.1 Sol ↑2 100.0
- 4 GPT-6.1 Terra ↑4 100.0
- 5
Gemini 4 Carbon ↑1 100.0
- 6 GPT-6.2 Astra ↓5 100.0
- 7 GPT-6.2 Sol ↓5 100.0
- 8
Gemini 4 Neon ↓4 100.0
- 9 GPT-6 Luna ↑1 46.2
- 10 Grok 4.7 ↓1 0.00
20-tick clock
10 models we track have a score here.
- 1 GPT-6 Astra ↑2 100.0
- 2 Grok 4.7 ↑7 100.0
- 3 Claude Opus 5.5 ↑4 100.0
- 4 GPT-6.1 Sol ↑1 100.0
- 5
Gemini 4 Carbon ↑1 100.0
- 6 GPT-6.2 Astra ↓5 100.0
- 7 GPT-6.2 Sol ↓5 100.0
- 8
Gemini 4 Neon ↓4 100.0
- 9 GPT-6.1 Terra ↓1 80.0
- 10 GPT-6 Luna 44.4
120-tick clock
10 models we track have a score here.
- 1 GPT-6.2 Astra 100.0
- 2 GPT-6.2 Sol 80.0
- 3 GPT-6 Astra 66.7
- 4 Claude Opus 5.5 ↑3 66.7
- 5
Gemini 4 Carbon ↑1 66.7
- 6
Gemini 4 Neon ↓2 66.7
- 7 Grok 4.7 ↑2 57.1
- 8 GPT-6.1 Sol ↓3 57.1
- 9 GPT-6.1 Terra ↓1 57.1
- 10 GPT-6 Luna 26.7
30-tick pulse
10 models we track have a score here.
- 1 GPT-6.2 Astra 100.0
- 2
Gemini 4 Carbon ↑4 90.0
- 3 GPT-6.2 Sol ↓1 90.0
- 4
Gemini 4 Neon 75.0
- 5 GPT-6.1 Sol 56.3
- 6 Claude Opus 5.5 ↑1 50.0
- 7 GPT-6 Astra ↓4 37.5
- 8 Grok 4.7 ↑1 37.5
- 9 GPT-6.1 Terra ↓1 37.5
- 10 GPT-6 Luna 18.8
More effort adds a median 6 score
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.
Source: Official RedstoneBench leaderboard
| Model | Effort | Harness | Score | Cost of the full run | Output tokens |
|---|---|---|---|---|---|
| GPT-6.2 Astra | max | 88.7 | |||
| GPT-6.2 Astra | ultra | 87.6 | |||
| GPT-6.2 Astra | xhigh | 82.0 | |||
| GPT-6.2 Astra | high | 81.5 | |||
| GPT-6.2 Sol | max | 75.8 | |||
| GPT-6.2 Astra | medium | 72.1 | |||
| GPT-6.2 Sol | ultra | 69.2 | |||
| GPT-6.2 Sol | high | 64.3 | |||
| GPT-6 Astra | ultra | 63.9 | |||
| high | 63.4 | ||||
| GPT-6.1 Sol | ultra | 62.5 | |||
| GPT-6.2 Astra | low | 61.4 | |||
| GPT-6.2 Sol | xhigh | 60.9 | |||
| xhigh | 60.3 | ||||
| GPT-6 Astra | max | 59.8 | |||
| GPT-6 Astra | xhigh | 59.7 | |||
| high | 59.4 | ||||
| GPT-6.1 Sol | xhigh | 59.3 | |||
| GPT-6.2 Sol | medium | 58.0 | |||
| GPT-6.1 Sol | max | 57.3 | |||
| Claude Opus 5.5 | medium | 57.1 | |||
| GPT-6 Astra | high | 54.4 | |||
| GPT-6.1 Sol | high | 50.3 | |||
| GPT-6.1 Terra | max | 49.5 | |||
| Claude Opus 5.5 | high | 49.5 | |||
| Claude Opus 5.5 | xhigh | 48.6 | |||
| GPT-6.1 Sol | medium | 48.1 | |||
| GPT-6.2 Sol | low | 48.1 | |||
| GPT-6 Astra | medium | 46.0 | |||
| GPT-6 Astra | low | 45.1 | |||
| Claude Opus 5.5 | low | 43.7 | |||
| GPT-6.1 Terra | ultra | 39.4 | |||
| GPT-6.1 Sol | low | 37.5 | |||
| GPT-6.1 Terra | xhigh | 36.4 | |||
| GPT-6.1 Terra | high | 33.8 | |||
| GPT-6.1 Terra | low | 32.6 | |||
| GPT-6.1 Terra | medium | 31.9 | |||
| Grok 4.7 | high | 29.2 | |||
| Claude Opus 5.5 | max | 28.1 | |||
| Grok 4.7 | medium | 27.2 | |||
| Grok 4.7 | xhigh | 21.6 | |||
| GPT-6 Luna | max | 19.2 | |||
| GPT-6 Luna | xhigh | 17.8 | |||
| Grok 4.7 | low | 16.1 | |||
| GPT-6 Luna | high | 8.47 | |||
| GPT-6 Luna | medium | 7.74 | |||
| GPT-6 Luna | low | 4.17 |
2 models have held number one on the official leaderboard since its release in November 2026
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official RedstoneBench leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official RedstoneBench 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
- 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
All 10 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
- Pass Independent runs Scores from the official leaderboard (10 entries).
- Pass Models scored 10 models we track.
- Pass Newest score The newest is from November 2026, for Gemini 4 Neon.
- Pass Frontier models 6 of the top 10 have a score, led by GPT-6.2 Astra (#1).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on redstone-bench.adamholter.chatgpt.site: 47 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.