SenseNova-U1-8B-MoT: benchmark results

Provider: SenseTime. Released 2026-04-27. Access: Open.

Aggregate Rating 1532 ± 47, rank #1827 of 10546 rated models, from 18 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
WeGenBench - Text Rendering74Character-level F1 (0-1 scaled to %) between the text render82.4
WeGenBench - Alignment93Checklist-based QA score (0-1 scaled to %): a Qwen3.5-397B-A58.8
Game2Real-Bench - Real World - Spatial Reasoning33Task score (%)58.3
Game2Real-Bench - Toy Game - Spatial Reasoning16.7Accuracy (%)52.1
WeGenBench - Alignment (COT Review)7.81COT-based deduction score (1-10): a fine-tuned expert vision47.1
Game2Real-Bench - Toy Game - Sequential Planning18Accuracy (%)39.6
Game2Real-Bench - Real World - Sequential Planning36Task score (%)37.5
Game2Real-Bench - Toy Game - Fine-grained Perception13.5Accuracy (%)37.5
WeGenBench - Text Rendering Sentence Accuracy52Sentence accuracy (0-1 scaled to %), the exact-match rate of35.3
GameHorizon - Offline T2 Multi-Horizon Decomposition59.2Accuracy (%)33.3
GameHorizon - Offline T3 Cross-Horizon Consistency67.5Accuracy (%)28.6
Game2Real-Bench - Overall18.5Mean per-sample score (%)25

Interactive version: theaggregate.ai/model?slug=sensenova-u1-8b-mot · How It Works · Data refreshed daily, snapshot 2026-11-18.