InternVL3.5-241B-A28B-Instruct: benchmark results
Provider: Shanghai AI Lab. Released 2025-08-26. Access: Open.
Aggregate Rating 1640 ± 23, rank #687 of 10916 rated models, from 26 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Omni-I2C (SMILES) - Tanimoto Similarity | 68.9 | Tanimoto similarity (0-100) between Morgan fingerprints of t | 91.7 |
| Video-IFBench - Nested | 10.4 | Task-gated instruction satisfaction rate (%; TISR on nested | 75.7 |
| Video-MME-v2 - Level 3 Complex Reasoning (w/o subs) | 14.05 | Level 3 Score w/o sub (%) | 70.8 |
| Video-MME-v2 | 32.9 | Avg Accuracy w/o sub (%) | 68.8 |
| Video-MME-v2 - Level 2 Temporal Understanding (w/o subs) | 16.31 | Level 2 Score w/o sub (%) | 68.8 |
| Video-MME-v2 - Non-Lin Score (w/o subs) | 15.8 | Non-Lin Score w/o sub (%) | 68.8 |
| Omni-I2C (LaTeX) - Visual Fidelity | 54.5 | Visual fidelity S_img (0-100): GPT-4.1 ratings of compositio | 66.7 |
| Omni-I2C (Python) - Visual Fidelity | 73.8 | Visual fidelity S_img (0-100): GPT-4.1 ratings of compositio | 66.7 |
| Video-MME-v2 - Level 1 Retrieval & Aggregation (w/o subs) | 18.12 | Level 1 Score w/o sub (%) | 66.7 |
| Omni-I2C (HTML) - Visual Fidelity | 76.6 | Visual fidelity S_img (0-100): GPT-4.1 ratings of compositio | 58.3 |
| Omni-I2C (Python) - Functional Coverage | 55.9 | Functional coverage S_func (0-100): share of the reference's | 58.3 |
| Omni-I2C - Visual Fidelity | 66.5 | Visual fidelity S_img (0-100): GPT-4.1 ratings of compositio | 58.3 |
Interactive version: theaggregate.ai/model?slug=internvl3-5-241b-a28b-instruct · How It Works · Data refreshed daily, snapshot 2026-11-08.