High-quality Vision-Language Evaluation on MMStar
68.2ScoreInternVL3
Evaluation Results
| Method | Links | |
|---|---|---|
| InternVL3LLM=Qwen2.5-7B, # Data=>6B / 100M / 22M, Model Scale=8B, Architecture Category=Modular, Reinforcement Learning=true2025.10 | 68.2 | |
| InternVL2.5LLM=InternLM2.5-7B, # Data=>6B / 50M / 4M, Model Scale=8B, Architecture Category=Modular, Reinforcement Learning=false2025.10 | 64.4 | |
| Qwen2.5-VLLLM=Qwen2.5-7B, # Data=- / - / -, Model Scale=8B, Architecture Category=Modular, Reinforcement Learning=true2025.10 | 63.9 | |
| Encoder-BasedLLM=Qwen3-8B, # Data=>6B / 40M / 4M, Model Scale=8B, Architecture Category=Modular, Reinforcement Learning=false2025.10 | 63.5 | |
| NEOLLM=Qwen3-8B, # Data=345M / 40M / 4M, Model Scale=8B, Architecture Category=Native, Reinforcement Learning=false2025.10 | 62.4 | |
| InternVL3LLM=Qwen2.5-1.5B, # Data=>6B / 100M / 22M, Model Scale=2B, Architecture Category=Modular, Reinforcement Learning=true2025.10 | 60.7 | |
| Qwen2-VLLLM=Qwen2-7B, # Data=- / - / -, Model Scale=8B, Architecture Category=Modular, Reinforcement Learning=false2025.10 | 60.7 | |
| Qwen2.5-VLLLM=Qwen2.5-3B, # Data=- / - / -, Model Scale=2B, Architecture Category=Modular, Reinforcement Learning=true2025.10 | 55.9 | |
| NEOLLM=Qwen3-1.7B, # Data=345M / 40M / 4M, Model Scale=2B, Architecture Category=Native, Reinforcement Learning=false2025.10 | 54.2 | |
| InternVL2.5LLM=InternLM2.5-1.8B, # Data=>6B / 100M / 16M, Model Scale=2B, Architecture Category=Modular, Reinforcement Learning=false2025.10 | 53.7 | |
| SAILLLM=Mistral-7B, # Data=512M / 86M / 6M, Model Scale=8B, Architecture Category=Native, Reinforcement Learning=false2025.10 | 53.1 | |
| Encoder-BasedLLM=Qwen3-1.7B, # Data=>6B / 40M / 4M, Model Scale=2B, Architecture Category=Modular, Reinforcement Learning=false2025.10 | 52.7 | |
| BREENLLM=Qwen2.5-7B, # Data=13M / 0M / 4M, Model Scale=8B, Architecture Category=Native, Reinforcement Learning=false2025.10 | 51.2 | |
| Qwen2-VLLLM=Qwen2-1.5B, # Data=- / - / -, Model Scale=2B, Architecture Category=Modular, Reinforcement Learning=false2025.10 | 48 |