Can the RTX 4050 Laptop 6 GB run MiniMax H3 Pruned?
Not entirely. Even Q4_K_M is 11.4 GB over 6 GB, so ComfyUI keeps part of it in system RAM.
The smallest sensible file, Q4_K_M (11.6 GB), needs about 17.4 GB - 11.4 GB more than RTX 4050 Laptop 6 GB has. ComfyUI can still run it by streaming part of the model from system RAM. How much slower that is depends on your ComfyUI version, the file format and the PCIe link - see the note below.
Also try the FP8 file (21.0 GB) on this GPU. It does not fit entirely, but a current ComfyUI streams the overflow from RAM and native 8-bit files often beat a small GGUF that does fit. Dynamic VRAM →
01With more VRAM
With 32 GB you could run MiniMax H3 Pruned at 8-bit or better (FP8, 21.0 GB) with no offloading: for example on the RTX 5090 32 GB.
02Every MiniMax H3 Pruned file on 6 GB
| File | Size | Needed | On this card | Quality | Download |
|---|---|---|---|---|---|
| 16-bit SAFETENSORS · Comfy-Org | 40.2 GB | 46.0 GB | 40.0 GB too big | the original weights | Hugging Face → |
| FP8 SAFETENSORS · Comfy-Org | 21.0 GB | 26.8 GB | 20.8 GB too big | practically identical to the original | Hugging Face → |
| INT8 SAFETENSORS · Comfy-Org | 21.0 GB | 26.8 GB | 20.8 GB too big | practically identical to the original | Hugging Face → |
| Q8_0 GGUF · Abiray | 21.6 GB | 27.4 GB | 21.4 GB too big | practically identical to the original | Hugging Face → |
| Q6_K GGUF · Abiray | 16.7 GB | 22.5 GB | 16.5 GB too big | very close to the original | Hugging Face → |
| Q5_K_M GGUF · Abiray | 14.1 GB | 19.9 GB | 13.9 GB too big | close; small differences in fine detail | Hugging Face → |
| Q4_K_M ← GGUF · Abiray | 11.6 GB | 17.4 GB | 11.4 GB too big | good; some loss in fine detail and text | Hugging Face → |
| Q3_K_M GGUF · Abiray | 8.9 GB | 14.7 GB | 8.7 GB too big | noticeable loss of detail | Hugging Face → |
“Needed” adds 5 GB of working memory for MiniMax H3 Pruned and a 0.8 GB reserve. Sizes 2026-10-01.
03What to download
| Part | File | Folder | Size | |
|---|---|---|---|---|
| Model | MiniMax-H3-FL2VA-Pruned-Q4_K_M.gguf Q4_K_M · Abiray/MiniMax-H3-Pruned-GGUF | models/unet | 11.6 GB | Download → |
| Text encoder | qwen3vl_32b_minimax_h3_int8_convrot.safetensors Qwen3-VL 32B INT8 (convrot) · Comfy-Org/MiniMax-H3 | models/text_encoders | 27.1 GB | Download → |
| VAE | minimax_h3_video_vae_fp16.safetensors MiniMax H3 video VAE FP16 · Comfy-Org/MiniMax-H3 | models/vae | 5.2 GB | Download → |
| Also needed | minimax_h3_audio_vae_fp32.safetensors MiniMax H3 audio VAE FP32 · Comfy-Org/MiniMax-H3 | models/vae | 0.6 GB | Download → |
| Total download · keep about the same free on disk | 44.5 GB | |||
Alternatives: Qwen3-VL 32B NVFP4 (AWQ) (15.7 GB); Qwen3-VL 32B BF16 (51.5 GB). FP4 files are listed only as alternatives unless the GPU is an RTX 50 (Blackwell). Sizes from Hugging Face, 2026-10-01.
System RAM: 64 GB or more. These files take about 44.5 GB of RAM while loaded, 50.5 GB with Windows and ComfyUI. What happens with less. calculated
04The text encoder
At 15.7 GB even the smallest Qwen3-VL 32B file is too big for 6 GB, so it runs from system RAM. That slows down reading the prompt, not the picture. Sizes: 16-bit 51.5 GB, INT8 27.1 GB, NVFP4 15.7 GB. Why the encoder does not count against VRAM →
05Measured and reported results
Nobody has sent measured numbers for this pair yet. If you run MiniMax H3 Pruned on a RTX 4050 Laptop 6 GB, send your time per image and peak VRAM and it will appear here, credited.
06About the RTX 4050 Laptop 6 GB
The RTX 4050 Laptop 6 GB is an Ada Lovelace GPU with hardware FP8, so ComfyUI can compute Comfy-Org's FP8 files natively: small and fast. (Plain FP8 files use FP8 maths with the --fast fp8_matrix_mult option.) As a laptop GPU it runs at a lower power limit than desktop cards (35–115 W depending on the laptop). The memory verdicts are the same; speed depends heavily on how much power the laptop maker allows.
6 GB GDDR6 · 96-bit · Ada Lovelace · 35–115 W. Everything that runs on it →
07Questions
How much VRAM does MiniMax H3 Pruned need?
Around 14.7 GB with the smallest file (Q3_K_M) and 26.8 GB with an 8-bit file (FP8). The full 16-bit file needs about 46.0 GB.
Which MiniMax H3 Pruned file should I download for the RTX 4050 Laptop 6 GB?
Q4_K_M (11.6 GB) from Abiray/MiniMax-H3-Pruned-GGUF. Load it with Unet Loader (GGUF) from the ComfyUI-GGUF node pack.
FP8 or Q8_0 for MiniMax H3 Pruned on the RTX 4050 Laptop 6 GB?
FP8 (21.0 GB): the RTX 4050 Laptop 6 GB computes it natively, while Q8_0 (21.6 GB) is unpacked on the fly.
How much do I need to download for MiniMax H3 Pruned on the RTX 4050 Laptop 6 GB?
44.5 GB: MiniMax-H3-FL2VA-Pruned-Q4_K_M.gguf, qwen3vl_32b_minimax_h3_int8_convrot.safetensors, minimax_h3_video_vae_fp16.safetensors, minimax_h3_audio_vae_fp32.safetensors.