family qwen · 15 variants · unknown
no scores.
| variant | kind | quant | bits | vRAM | dl | org | year |
|---|---|---|---|---|---|---|---|
| Qwen3-VL-4B-Thinking | original | bf16 | 16.0 | 23,933 | Qwen | 2025 | |
| Qwen3-VL-4B-Thinking-unsloth-bnb-4bit | bnb-4bit | 4bit | 19,033 | unsloth | 2025 | ||
| Qwen3-VL-4B-Thinking-GGUF | gguf | gguf | 16.0 | 3,665 | unsloth | 2025 | |
| Qwen3-VL-4B-Thinking-FP8 | fp8 | fp8 | 8.0 | 3,506 | Qwen | 2025 | |
| Qwen3-VL-4B-Thinking-GGUF | gguf | gguf | 16.0 | 2,359 | Qwen | 2025 | |
| Qwen3-VL-4B-Thinking-1M-GGUF | gguf | gguf | 16.0 | 1,289 | unsloth | 2025 | |
| Qwen_Qwen3-VL-4B-Thinking-GGUF | gguf | gguf | 2.0 | 971 | bartowski | 2025 | |
| Qwen3-VL-4B-Thinking-GGUF | gguf | gguf | 4.0 | 953 | lmstudio-community | 2025 | |
| EGM-4B-SFT | original | bf16 | 16.0 | 750 | nvidia | 2026 | |
| EGM-4B | original | bf16 | 16.0 | 158 | nvidia | 2026 | |
| Qwen3-VL-4B-Thinking-MLX-4bit | mlx | mlx | 95 | lmstudio-community | 2025 | ||
| Qwen3-VL-4B-Thinking | original | bf16 | 16.0 | 74 | unsloth | 2025 | |
| Qwen3-VL-4B-Thinking-MLX-8bit | mlx | mlx | 71 | lmstudio-community | 2025 | ||
| Qwen3-VL-4B-Thinking-bnb-4bit | bnb-4bit | 4bit | 66 | unsloth | 2025 | ||
| Qwen3-VL-4B-Thinking-MLX-5bit | mlx | mlx | 54 | lmstudio-community | 2025 |