HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/Qwen/Qwen/Qwen2-VL-7B-Instruct

Qwen/Qwen2-VL-7B-Instruct

family qwen · 12 variants · unknown

provider
Qwen
released
—
huggingface
https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct
best scores
MMMU: 53.77DocVQA: 93.89MMBench: 81.78MathVista: 58.3ChartQA: 83.3MT-Bench: 63.7

Best scores per benchmark

benchmarkcategorybest score
DocVQAmultimodal93.89
ChartQAmultimodal83.3
MMBenchmultimodal81.78
MT-Benchchat63.7
MathVistamultimodal58.3
MMMUmultimodal53.77

Variants (12)

variantkindquantbitsvRAMdlorgyear
Qwen2-VL-7B-Instruct-AWQawqawq21,809,222Qwen2024
Qwen2-VL-7B-Instruct-GGUFggufgguf2.06,687bartowski2024
Qwen2-VL-7B-Instruct-unsloth-bnb-4bitbnb-4bit4bit5,713unsloth2025
olmOCR-7B-0225-previeworiginalbf1616.04,084allenai2025
Qwen2-VL-7B-Instruct-GPTQ-Int8gptqgptq2,811Qwen2024
Qwen2-VL-7B-Instruct-GGUFggufgguf3.02,210lmstudio-community2025
Qwen2-VL-7B-Instruct-GPTQ-Int4gptqgptq2,195Qwen2024
Qwen2-VL-7B-Instruct-bnb-4bitbnb-4bit4bit1,703unsloth2024
Qwen2-VL-7B-Instruct-GGUFggufgguf4.01,013ggml-org2025
Qwen2-VL-7B-Instructoriginalbf1616.0250unsloth2025
GUI-Actor-7B-Qwen2-VLoriginalbf1616.0154microsoft2025
olmOCR-7B-0225-preview-GGUFggufgguf16.093Mungert2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.