Multimodal Understanding on SEED-2-Plus
81.1AccuracyVLSI-2B
Evaluation results
| Method | Links | ||
|---|---|---|---|
| VLSI-2BSize=2B2024.12 | 81.1 | — | |
| Qwen2.5-VL-7B-MastersSize=7B2025.12 | 75.9 | — | |
| MastersBase Model=Qwen2.5-VL-7B2025.12 | 75.9 | — | |
| InternVL3.5-8B-MastersSize=8B2025.12 | 75.5 | — | |
| MastersBase Model=InternVL3.5-8B2025.12 | 75.5 | — | |
| Qwen3-VL-8B-MastersSize=8B2025.12 | 75.1 | — | |
| MastersBase Model=Qwen3-VL-8B2025.12 | 75.1 | — | |
| InternVL3-8B-MastersSize=8B2025.12 | 75 | — | |
| MastersBase Model=InternVL3-8B2025.12 | 75 | — | |
| VLSI-7BSize=7B2024.12 | 74.9 | — | |
| GLM-4.5V2025.12 | 74 | — | |
| GPT-4.12025.12 | 73.1 | — | |
| Qwen2.5-VL-72B2025.12 | 73 | — | |
| Qwen3-VL-4B-MastersSize=4B2025.12 | 72.9 | — | |
| Qwen3-VL-32B2025.12 | 72.8 | — | |
| MiMo-VL-8BSize=8B2025.12 | 72.4 | — | |
| Qwen2-VL-72BSize=72B2024.12 | 72.3 | — | |
| Qwen2-VL-72B2025.06 | 72.3 | — | |
| InternVL2.5-8B-GenRecalTeacher VLM=InternVL2.5-78B2025.06 | 72.3 | — | |
| Qwen2.5-VL-3B-MastersSize=3B2025.12 | 72.1 | — | |
| GPT-4o2025.12 | 72 | — | |
| GPT-4o (0513)2025.06 | 72 | — | |
| InternVL3-78B2025.12 | 71.9 | — | |
| GLM-4.1V-9BSize=9B2025.12 | 71.8 | — | |
| Claude3.5-Sonnet2024.12 | 71.7 | — | |
| Claude-3.5-Sonnet2025.12 | 71.7 | — | |
| Claude-3.5-Sonnet2025.06 | 71.7 | — | |
| InternVL3.5-2B-MastersSize=2B2025.12 | 71.6 | — | |
| InternVL3.5-4B-MastersSize=4B2025.12 | 71.4 | — | |
| InternVL2.5-78B2025.06 | 71.3 | — | |
| InternVL2.5-8B-GenRecalTeacher VLM=Qwen2-VL-72B2025.06 | 71.3 | — | |
| InternVL3.5-38B2025.12 | 71 | — | |
| InternVL2.5-8B-GenRecalTeacher VLM=InternVL2-76B2025.06 | 70.9 | — | |
| Gemini-1.5-Pro2024.12 | 70.8 | — | |
| GPT-4o (0806)Model Version=08062024.12 | 70.8 | — | |
| Qwen3-VL-8BSize=8B2025.12 | 70.8 | — | |
| InternVL3.5-8BSize=8B2025.12 | 70.8 | — | |
| Gemini-1.5-Pro2025.12 | 70.8 | — | |
| Gemini-1.5-Pro2025.06 | 70.8 | — | |
| Ovis2-8BSize=8B2025.12 | 70.4 | — | |
| Qwen2.5-VL-7BSize=7B2025.12 | 70.4 | — | |
| Qwen3-VL-4BSize=4B2025.12 | 70.2 | — | |
| InternVL2-76BSize=76B2024.12 | 70 | — | |
| InternVL3-8BSize=8B2025.12 | 69.7 | — | |
| InternVL2-76B2025.06 | 69.7 | — | |
| fp16Model=Qwen2-VL-7b, Bitwidth=fp162026.02 | 69.4 | — | |
| InternVL3.5-4BSize=4B2025.12 | 69.4 | — | |
| Qwen3-VL-2B-MastersSize=2B2025.12 | 69.4 | — | |
| GPT-4V (0409)Model Version=04092024.12 | 69.3 | — | |
| Ovis2-4BSize=4B2025.12 | 69.3 | — | |
| InternVL2.5-8B-GenRecalTeacher VLM=NVLM-72B2025.06 | 69.3 | — | |
| LLaVA-OneVision-1.5-8BSize=8B2025.12 | 69.2 | — | |
| LLaVA-OneVision-1.5-4BSize=4B2025.12 | 68.9 | — | |
| Phi-4-Multimodal-5.6BSize=5.6B2025.12 | 68.5 | — | |
| NVLM-D-72BSize=72B2024.12 | 68.4 | — | |
| NVLM-72B2025.12 | 68.4 | — | |
| NVLM-72B2025.06 | 68.4 | — | |
| LLaMA-3.2-Vision2024.12 | 68.2 | — | |
| InternVL3.5-2BSize=2B2025.12 | 68 | — | |
| MiniCPM-o2.6-8BSize=8B2025.12 | 67.9 | — | |
| Keye-VL-8BSize=8B2025.12 | 67.8 | — | |
| InternVL3-2B-MastersSize=2B2025.12 | 67.8 | — | |
| Qwen2.5-VL-3BSize=3B2025.12 | 67.6 | — | |
| Claude-3.7-Sonnet2025.12 | 67.6 | — | |
| Molmo-72B2025.12 | 67.6 | — | |
| Ovis2-2BSize=2B2025.12 | 67.4 | — | |
| RTNModel=Qwen2-VL-7b, Bitwidth=W4A82026.02 | 66.5 | — | |
| MBQModel=Qwen2-VL-7b, Bitwidth=W4A82026.02 | 66.5 | — | |
| Qwen3-VL-2BSize=2B2025.12 | 66.4 | — | |
| TLQModel=Qwen2-VL-7b, Bitwidth=W4A82026.02 | 66.3 | — | |
| MiniCPM-V2.6-8BSize=8B2025.12 | 65.7 | — | |
| Phantom-7BSize=7B2024.12 | 65.5 | — | |
| LLaVA-OneVision-7BSize=7B2025.12 | 65.4 | — | |
| SmoothQuantModel=Qwen2-VL-7b, Bitwidth=W4A82026.02 | 65.3 | — | |
| Cambrian-34BSize=34B2024.12 | 65.1 | — | |
| fp16Model=LLaVA-onevision-7b, Bitwidth=fp162026.02 | 64.8 | — | |
| InternVL3-2BSize=2B2025.12 | 64.6 | — | |
| TLQModel=Qwen2-VL-7b, Bitwidth=W4A62026.02 | 64.3 | — | |
| CapRL-Image-5MModel Configuration=InternLM2.5-7B + CLIP-ViT-L2026.06 | 63.5 | — | |
| SmoothQuantModel=Qwen2-VL-7b, Bitwidth=W4A62026.02 | 63.2 | — | |
| CapRL-Image-5MModel Configuration=Qwen2.5-3B + Qwen2.5-ViT2026.06 | 63.2 | — | |
| CapRL-Image-5MModel Configuration=Qwen2.5-7B + Qwen2.5-ViT2026.06 | 63.2 | — | |
| CapRL-Image-1MModel Configuration=Qwen2.5-7B + Qwen2.5-ViT2026.06 | 63.1 | — | |
| Aquila-VL-2BSize=2B2025.12 | 63 | — | |
| TLQModel=LLaVA-onevision-7b, Bitwidth=W4A82026.02 | 62.8 | — | |
| CapRL-Image-1MModel Configuration=InternLM2.5-7B + CLIP-ViT-L2026.06 | 62.8 | — | |
| RTNModel=Qwen2-VL-7b, Bitwidth=W4A62026.02 | 62.7 | — | |
| RTNModel=LLaVA-onevision-7b, Bitwidth=W4A82026.02 | 62.4 | — | |
| MBQModel=LLaVA-onevision-7b, Bitwidth=W4A82026.02 | 62.4 | — | |
| fp16Model=Qwen2-VL-2b, Bitwidth=fp162026.02 | 62.3 | — | |
| MBQModel=Qwen2-VL-7b, Bitwidth=W4A62026.02 | 62.3 | — | |
| Phi-3.5-Vision-4BSize=4B2025.12 | 62.2 | — | |
| CapRL-Image-1MModel Configuration=Qwen2.5-3B + Qwen2.5-ViT2026.06 | 62 | — | |
| SmoothQuantModel=LLaVA-onevision-7b, Bitwidth=W4A82026.02 | 61.4 | — | |
| DenseFusion-1MModel Configuration=Qwen2.5-7B + Qwen2.5-ViT2026.06 | 61 | — | |
| ShareGPT4V-1MModel Configuration=Qwen2.5-7B + Qwen2.5-ViT2026.06 | 60.9 | — | |
| TLQModel=LLaVA-onevision-7b, Bitwidth=W4A62026.02 | 60.7 | — | |
| ShareGPT4V-1MModel Configuration=Qwen2.5-3B + Qwen2.5-ViT2026.06 | 60.5 | — | |
| RTNModel=LLaVA-onevision-7b, Bitwidth=W4A62026.02 | 60.4 | — | |
| VanillaModel Configuration=Qwen2.5-7B + Qwen2.5-ViT2026.06 | 60.4 | — |