HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/Qwen/Qwen/Qwen3-VL-8B-Instruct

Qwen/Qwen3-VL-8B-Instruct

family qwen · 24 variants · unknown

provider
Qwen
released
—
huggingface
https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct
best scores
none

Best scores per benchmark

no scores.

Variants (24)

variantkindquantbitsvRAMdlorgyear
Qwen3-VL-8B-Instructoriginalbf1616.05,122,647Qwen2025
Qwen3-VL-8B-Instruct-FP8fp8fp88.02,716,164Qwen2025
Qwen3-VL-Embedding-8Boriginalbf1616.082,193,134Qwen2026
Cosmos-Reason2-8Boriginalbf1616.0272,459nvidia2026
Qwen3-VL-8B-Instruct-MLX-4bitmlxmlx93,137lmstudio-community2025
Qwen3-VL-Reranker-8Boriginalbf1616.0889,583Qwen2026
Qwen3-VL-8B-Instruct-MLX-8bitmlxmlx78,640lmstudio-community2025
Qwen3-VL-8B-Instruct-MLX-5bitmlxmlx77,587lmstudio-community2025
Qwen3-VL-8B-Instruct-MLX-6bitmlxmlx77,430lmstudio-community2025
Qwen3-VL-8B-Instruct-GGUFggufgguf4.067,315lmstudio-community2025
Qwen3-VL-8B-Instruct-GGUFggufgguf16.062,282Qwen2025
Qwen3-VL-8B-Instruct-GGUFggufgguf16.030,245unsloth2025
Qwen3-VL-8B-Instruct-unsloth-bnb-4bitbnb-4bit4bit23,110unsloth2025
Qwen3-VL-8B-Instruct-bnb-4bitbnb-4bit4bit13,270unsloth2025
Qwen_Qwen3-VL-8B-Instruct-GGUFggufgguf2.04,167bartowski2025
Qwen3-VL-8B-Instructoriginalbf1616.04,067unsloth2025
SingGuard-0.8boriginalbf1616.01,058inclusionAI2026
Qwen3-VL-8B-Instruct-GGUFggufgguf4.01,052mradermacher2025
Qwen3-VL-8B-Instruct-1M-GGUFggufgguf16.0807unsloth2025
SingGuard-8boriginalbf1616.0652inclusionAI2026
SingGuard-8b-GGUFggufgguf16.0533inclusionAI2026
Qwen3-VL-8B-Instruct-GGUFggufgguf16.0413Mungert2025
ZwZ-8Boriginalbf1616.0227inclusionAI2026
Visual-ERMoriginalbf1616.029internlm2026
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.