InternVL3.5-241B: benchmark results
Provider: Shanghai AI Lab.
Aggregate Rating 1649 ± 36, rank #636 of 10916 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ReceiptBench - Normalization | 87.91 | Formatting and normalization fields (dates as YYYY-MM-DD, to | 84.6 |
| ReceiptBench - Structure | 51.12 | Structural parsing field (the structured list of line items | 84.6 |
| ReceiptBench - Perception | 68.53 | Basic perception fields (raw times, totals, currency clues, | 76.9 |
| ReceiptBench | 67.42 | Overall field-level F1 (times 100, so 0-100) on the 2,000-im | 69.2 |
| ReceiptBench - Reasoning | 80.24 | Semantic reasoning fields (expense type, place, departure, a | 61.5 |
| ALL Bench Multimodal | 77.7 | Average Numeric VLM Score (%) | 53.3 |
| Pause-and-Think-B | 50.67 | Binary validity accuracy (%): the GPT-5.1 judge marks each f | 50 |
| Pause-and-Think-B - Goal Planning | 66.67 | Binary validity accuracy (%) on the 200 goal-planning items | 40 |
| Pause-and-Think-B - Scene Understanding | 43.81 | Binary validity accuracy (%) on the 100 scene-understanding | 30 |
| BabyVision | 19.2 | Score (%) | 20.8 |
Interactive version: theaggregate.ai/model?slug=internvl3-5-241b · How It Works · Data refreshed daily, snapshot 2026-11-08.