family qwen · 7 variants · unknown
| benchmark | category | best score |
|---|---|---|
| DocVQA | multimodal | 95.79 |
| MMBench | multimodal | 86.94 |
| ChartQA | multimodal | 76.2 |
| MathVista | multimodal | 70.19 |
| MMMU | multimodal | 65.44 |
| variant | kind | quant | bits | vRAM | dl | org | year |
|---|---|---|---|---|---|---|---|
| Qwen2-VL-72B-Instruct-AWQ | awq | awq | 64,535 | Qwen | 2024 | ||
| Qwen2-VL-72B-Instruct-GGUF | gguf | gguf | 2.0 | 1,337 | bartowski | 2024 | |
| Qwen2-VL-72B-Instruct-GPTQ-Int4 | gptq | gptq | 554 | Qwen | 2024 | ||
| Qwen2-VL-72B-Instruct-GGUF | gguf | gguf | 3.0 | 384 | lmstudio-community | 2025 | |
| Qwen2-VL-72B-Instruct-bnb-4bit | bnb-4bit | 4bit | 106 | unsloth | 2025 | ||
| Qwen2-VL-72B-Instruct-GPTQ-Int8 | gptq | gptq | 90 | Qwen | 2024 | ||
| Qwen2-VL-72B-Instruct | original | bf16 | 16.0 | 62 | unsloth | 2025 |