Multimodal Understanding on MMMU-Pro 1.0 (test)
44.39Overall ScoreVanilla (Full Tokens)
Evaluation Results
| Method | Links | |
|---|---|---|
| Vanilla (Full Tokens)Model=InternVL3.5-14B, Model Category=Large-scale MLLMs, Token Retention Ratio=100%2026.04 | 44.39 | |
| Vanilla (Full Tokens)Model=Qwen3-VL-8B, Model Category=Large-scale MLLMs, Token Retention Ratio=100%2026.04 | 42.31 | |
| DSTPModel=InternVL3.5-14B, Model Category=Large-scale MLLMs, Token Retention Ratio=33.3%2026.04 | 40.64 | |
| Vanilla (Full Tokens)Model=Qwen3-VL-4B-Thinking, Model Category=Thinking-based MLLMs, Token Retention Ratio=100%2026.04 | 40.51 | |
| Vanilla (Full Tokens)Model=InternVL3.5-8B-Thinking, Model Category=Thinking-based MLLMs, Token Retention Ratio=100%2026.04 | 37.68 | |
| DSTPModel=Qwen3-VL-4B-Thinking, Model Category=Thinking-based MLLMs, Token Retention Ratio=33.3%2026.04 | 36.52 | |
| DSTPModel=Qwen3-VL-8B, Model Category=Large-scale MLLMs, Token Retention Ratio=33.3%2026.04 | 33.31 | |
| FastV (33.3%)Model=InternVL3.5-14B, Model Category=Large-scale MLLMs, Token Retention Ratio=33.3%2026.04 | 31.66 | |
| DSTPModel=InternVL3.5-8B-Thinking, Model Category=Thinking-based MLLMs, Token Retention Ratio=33.3%2026.04 | 31.4 | |
| FastV (33.3%)Model=Qwen3-VL-4B-Thinking, Model Category=Thinking-based MLLMs, Token Retention Ratio=33.3%2026.04 | 29.77 | |
| FastV (33.3%)Model=InternVL3.5-8B-Thinking, Model Category=Thinking-based MLLMs, Token Retention Ratio=33.3%2026.04 | 20.63 | |
| FastV (33.3%)Model=Qwen3-VL-8B, Model Category=Large-scale MLLMs, Token Retention Ratio=33.3%2026.04 | 18.64 |