Furniture Assembly leaderboard
As of 16 Nov 2026, the top score on Furniture Assembly is 90.0% by Claude Fable 5.2 (Anthropic), ahead of Claude Opus 5.5 at 83.3%, across 39 model configurations with a published result (Epoch AI Benchmarking Hub, result dated 9 Nov 2026). Spatial reasoning on photos of IKEA furniture builds, some with a deliberate assembly mistake: say whether the build is right and, if not, which step went wrong. Random guessing scores about 30%. Run by Epoch AI. Furniture Assembly is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 16 Nov 2026)
- Claude Fable 5.2 holds the top Furniture Assembly score at 90.0% (Epoch AI Benchmarking Hub, 9 Nov 2026).
- Claude Opus 5.5 is second on Furniture Assembly at 83.3% (Epoch AI Benchmarking Hub, 22 Sep 2026).
- Furniture Assembly: Spatial reasoning on photos of IKEA furniture builds, some with a deliberate assembly mistake: say whether the build is right and, if not, which step went wrong. Random guessing scores about 30%. Run by Epoch AI (Epoch AI Benchmarking Hub).
- 39 model configurations have a published Furniture Assembly result on BenchLeader (BenchLeader, 16 Nov 2026).
- Furniture Assembly does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
Spatial reasoning on photos of IKEA furniture builds, some with a deliberate assembly mistake: say whether the build is right and, if not, which step went wrong. Random guessing scores about 30%. Run by Epoch AI.
- Published by
- Epoch AI Benchmarking Hub
- Category
- Multimodal
- Index weight
- Reference only
- Models
- 39
- Data as of
- 16 Nov 2026
- Newest result
- 10 Nov 2026
CC BY 4.0 — Epoch AI, ‘AI Benchmarking Hub’, epoch.ai/benchmarks. Mirrored boards credit their original publishers.
- 1Claude Fable 5.2 (max)90.0%
- 2Claude Opus 5.5 (max)83.3%
- 3GPT-6 Astra (max)80.0%
- 4GPT-6.1 Sol (max)80.0%
- 5GPT-6.1 Terra (max)79.2%
- 6Gemini 4 Carbon (high)77.5%
- 7Claude Sonnet 5.5 (max)75.0%
- 8Gemini 4 Argon (high)72.5%
- 9Claude Fable 5.1 (max)70.0%
- 10Claude Opus 5 (max)60.8%
- 11GPT-6 Sol (max)58.3%
- 12GPT-5.6 Sol (max)56.7%
- 13Kimi K3.1 Pro Together (max)55.8%
- 14Muse Wonder (max)54.2%
- 15GPT-5.6 Terra (max)54.2%
| # | ||||
|---|---|---|---|---|
| 1 | New | 90.0% | 72.7 | 2026-11-09 |
| 2 | 83.3% | 68.4 | 2026-09-22 | |
| 3 | 80.0% | 67.3 | 2026-09-03 | |
| 4 | 80.0% | 66.6 | 2026-09-29 | |
| 5 | 79.2% | 64.1 | 2026-10-26 | |
| 6 | 77.5% | 68.9 | 2026-11-10 | |
| 7 | 75.0% | 64.2 | 2026-09-28 | |
| 8 | 72.5% | 65.3 | 2026-10-13 | |
| 9 | 70.0% | 66.7 | 2026-09-01 | |
| 10 | 60.8% | 65.0 | 2026-07-24 | |
| 11 | 58.3% | 62.6 | 2026-09-22 | |
| 12 | 56.7% | 65.3 | 2026-07-09 | |
| 13 | 55.8% | 60.0 | 2026-10-23 | |
| 14 | 54.2% | 60.8 | 2026-10-12 | |
| 15 | 54.2% | 59.5 | 2026-07-09 | |
| 16 | 47.5% | 55.3 | 2026-10-07 | |
| 17 | 44.2% | 64.1 | 2026-04-23 | |
| 18 | 44.2% | 55.1 | 2026-09-22 | |
| 19 | 42.5% | 61.1 | 2026-05-28 | |
| 20 | 42.5% | 56.2 | 2026-07-09 | |
| 21 | 40.0% | 56.1 | 2026-08-12 | |
| 22 | 38.3% | 59.9 | 2025-12-11 | |
| 23 | 37.5% | 63.1 | 2026-03-05 | |
| 24 | 35.8% | 65.1 | 2026-06-09 | |
| 25 | 34.2% | 60.2 | 2026-07-16 | |
| 26 | 34.2% | 58.5 | 2026-09-09 | |
| 27 | 33.3% | 59.7 | 2026-04-16 | |
| 28 | 31.7% | 59.4 | 2026-09-02 | |
| 29 | 30.8% | – | 2026-10-29 | |
| 30 | 28.3% | 57.1 | 2025-11-24 | |
| 31 | 28.3% | 57.0 | 2026-02-05 | |
| 32 | 26.7% | 62.1 | 2026-08-13 | |
| 33 | 26.7% | 60.4 | 2026-02-19 | |
| 34 | 23.3% | 61.0 | 2026-07-21 | |
| 35 | 22.5% | 56.8 | 2026-07-08 | |
| 36 | 21.7% | 55.1 | 2026-04-20 | |
| 37 | 20.8% | 57.4 | 2026-09-21 | |
| 38 | 20.8% | 48.3 | 2026-10-06 | |
| 39 | 20.0% | 56.6 | 2026-09-01 |
Multimodal AI benchmark leaderboard · All LLM benchmarks
Related benchmarks
Top head-to-heads on Furniture Assembly
Cite as: BenchLeader, “Furniture Assembly leaderboard”, data as of 16 Nov 2026, https://www.benchleader.com/benchmarks/furniture_assembly.
Furniture Assembly: questions
- What does Furniture Assembly measure?
- Spatial reasoning on photos of IKEA furniture builds, some with a deliberate assembly mistake: say whether the build is right and, if not, which step went wrong. Random guessing scores about 30%. Run by Epoch AI. Scores are reported in percent of tasks solved; higher is better.
- What is the top score on Furniture Assembly?
- As of 16 Nov 2026, the highest score on Furniture Assembly is 90.0% by Claude Fable 5.2 (result dated 9 Nov 2026), ahead of Claude Opus 5.5 at 83.3%. Scores come from Epoch AI Benchmarking Hub and are re-read daily.
- Which AI model leads Furniture Assembly?
- Claude Fable 5.2 leads Furniture Assembly with 90.0% as of 16 Nov 2026 (result dated 9 Nov 2026), ahead of Claude Opus 5.5 at 83.3%.
- How many models have Furniture Assembly results?
- 39 model configurations have a Furniture Assembly result on BenchLeader, all taken from Epoch AI Benchmarking Hub.
- Who runs Furniture Assembly and how often is it updated?
- Furniture Assembly is published by Epoch AI Benchmarking Hub. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does Furniture Assembly count toward the BenchLeader Index?
- No. Furniture Assembly is shown for reference but left out of the composite index.