M-LongDoc (multimodal long-document benchmark) — Identical stats to M-LongDoc_acc -- same data, _acc is just the metric-name suffix.
| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | GPT-4o | OpenAI | 41.4 | 2 | 2026-06-12 |
| 2 | MiniMax VL 01 | MiniMax | 32.5 | 1 | 2026-06-12 |
| 3 | Gemini 2.0 Flash Exp | 31.4 | 1 | 2026-06-12 | |
| 4 | Claude 3.5 Sonnet | Anthropic | 31.4 | 2 | 2026-06-12 |
| 5 | Gemini 1.5 Pro | 26.2 | 2 | 2026-06-12 | |
| 6 | InternVL2.5-78B | OpenGVLab | 19.7 | 1 | 2026-06-12 |
| 7 | Llama 3.2 Instruct 90B (Vision) | Meta | 13.9 | 1 | 2026-06-12 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.