IKEA-Bench (Text Only) - Step Recognition: leaderboard

Metric: Accuracy (%) of four-way multiple-choice step recognition (match the video frames of one assembly step to one of four wordless manual diagrams, distractors from adjacent steps; 320 questions; chance 25) on IKEA-Bench, wordless IKEA manual diagrams against real assembly video from 29 products, Text Only setting (diagram images replaced by text descriptions), zero-shot greedy decoding, unparsed answers counted wrong; higher is better. Source: arxiv.org. Saturation forecast: Around September 2027. 17 models tracked.

Top models

#ModelScore
1Qwen 3.5 27B54.4
2Qwen 3.5 9B45.9
3GLM-4.1V-9B (Thinking)43.1
4Gemma 3 27B42.2
5Qwen 2.5 VL 7B39.4
6Gemma 3 12B36.9
7Qwen 3.5 2B33.1
8Gemma 3 4B32.2

Interactive version: theaggregate.ai/benchmark?slug=ikea-bench-text-only-step-recognition · How It Works · Data refreshed daily, snapshot 2026-10-07.