HunyuanVideo 1.5 on 8 GB of VRAM
Not entirely. Even Q4_K_M is 1.4 GB over 8 GB, so ComfyUI keeps part of it in system RAM. Tencent's 8.3B video model from late 2025, much lighter than the original 13B HunyuanVideo. The FP8 file is the cfg-distilled variant.
Nothing fits entirely, but Q4_K_M (5.1 GB) overflows by only about 1.4 GB. ComfyUI keeps that part in system RAM automatically: slower than a full fit, but usable.
Also try the FP8 file (8.3 GB) on NVIDIA GPUs. 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 →
01Other amounts of VRAM
02Every file, on 8 GB
| File | Size | Needed | On this card | Quality | Download |
|---|---|---|---|---|---|
| 16-bit SAFETENSORS · Comfy-Org | 16.7 GB | 21.0 GB | 13.0 GB too big | the original weights | Hugging Face → |
| Q8_0 GGUF · jayn7 | 9.0 GB | 13.3 GB | 5.3 GB too big | practically identical to the original | Hugging Face → |
| FP8 SAFETENSORS · Comfy-Org | 8.3 GB | 12.6 GB | 4.6 GB too big | practically identical to the original | Hugging Face → |
| Q6_K GGUF · jayn7 | 7.0 GB | 11.3 GB | 3.3 GB too big | very close to the original | Hugging Face → |
| Q5_K_M GGUF · jayn7 | 6.1 GB | 10.4 GB | 2.4 GB too big | close; small differences in fine detail | Hugging Face → |
| Q4_K_M ← GGUF · jayn7 | 5.1 GB | 9.4 GB | spills 1.4 GB | good; some loss in fine detail and text | Hugging Face → |
“Needed” adds 3.5 GB of working memory for HunyuanVideo 1.5 and a 0.8 GB reserve. Sizes 2026-10-01.
03What to download
| Part | File | Folder | Size | |
|---|---|---|---|---|
| Model | hunyuanvideo1.5_720p_t2v-Q4_K_M.gguf Q4_K_M · jayn7/HunyuanVideo-1.5_T2V_720p-GGUF | models/unet | 5.1 GB | Download → |
| Text encoder | qwen_2.5_vl_7b_fp8_scaled.safetensors Qwen2.5-VL 7B FP8 (scaled) · Comfy-Org/HunyuanVideo_1.5_repackaged | models/text_encoders | 9.4 GB | Download → |
| Text encoder | byt5_small_glyphxl_fp16.safetensors ByT5 small GlyphXL FP16 · Comfy-Org/HunyuanVideo_1.5_repackaged | models/text_encoders | 0.4 GB | Download → |
| VAE | hunyuanvideo15_vae_fp16.safetensors HunyuanVideo 1.5 VAE FP16 · Comfy-Org/HunyuanVideo_1.5_repackaged | models/vae | 2.5 GB | Download → |
| Total download · keep about the same free on disk | 17.4 GB | |||
Alternatives: Qwen2.5-VL 7B BF16 (16.6 GB). Only for some workflows: SigCLIP vision patch14 384 (0.9 GB, I2V only). Sizes from Hugging Face, 2026-10-01.
Free ComfyUI workflow, tested on my RTX 5060 Ti: 480p text to video, FP8 workflow (.json). Drag the file into ComfyUI. All workflows →
04The text encoder
At 9.4 GB even the smallest Qwen2.5-VL 7B + glyph encoder file is too big for 8 GB, so it runs from system RAM. That slows down reading the prompt, not the picture. Sizes: 16-bit 16.6 GB, FP8 9.4 GB. Why the encoder does not count against VRAM →