Hunyuan / VIDEO
HunyuanVideo T2V 720p VRAM requirements
Explore memory tradeoffs with sourced model data and transparent estimates.
FP8 estimate
46.4 GiB
at default settings
FP16 estimate
46.4 GiB
at default settings
Suggested GPU
1 × RTX A6000 48 GiB
Model files & facts
| Media | video |
|---|---|
| Default resolution | 1280 × 720 |
| Default frames | 129 |
| Temporal compression | 4 |
| Notes | Component-based diffusion memory estimate. |
| Component | Weight file (source) | Size |
|---|---|---|
| Backbone FP16 | split_files/diffusion_models/hunyuan_video_t2v_720p_bf16.safetensors | 23.88 GiB |
| CLIP-L FP16 | split_files/text_encoders/clip_l.safetensors | 0.23 GiB |
| LLaVA Llama 3 FP16 | split_files/text_encoders/llava_llama3_fp16.safetensors | 14.97 GiB |
| LLaVA Llama 3 FP8 | split_files/text_encoders/llava_llama3_fp8_scaled.safetensors | 8.47 GiB |
| VAE | split_files/vae/hunyuan_video_vae_bf16.safetensors | 0.46 GiB |
Precision & offloading comparison
Source default resolution and frame count · batch 1 · FP16 text encoders · safety margin included.
| Backbone | Text encoders | Total |
|---|---|---|
| FP16 | GPU | 46.35 GiB |
| FP16 | CPU | 46.35 GiB |
ComfyUI file placement
Example component locations; install matching loader nodes. Brace patterns in source filenames denote all shards. Preserve the source layout when a loader requires it.
models/diffusion_models/split_files/diffusion_models/hunyuan_video_t2v_720p_bf16.safetensors
models/text_encoders/split_files/text_encoders/clip_l.safetensors
models/text_encoders/split_files/text_encoders/llava_llama3_fp16.safetensors
models/text_encoders/split_files/text_encoders/llava_llama3_fp8_scaled.safetensors
models/vae/split_files/vae/hunyuan_video_vae_bf16.safetensorsRecommended GPUs by memory fit
FP16 backbone at source default resolution, FP16 text encoders on GPU. Smallest nominal memory first; green results only. Check backend and precision support separately.
1 × RTX A6000 48 GiB
Fits with safety headroom
1 × L40S 48 GiB
Fits with safety headroom
1 × NVIDIA A40 (48 GiB)
Fits with safety headroom
Sources & verification
Exact file sources are linked in the component table above. ComfyUI memory planner constant
Last verified: 2026-11-20
Try your configuration
Your workload
Custom diffusion model parameters
Uses the ComfyUI activation and decode profile for HunyuanVideo T2V 720p. Leave fields empty to use catalog files. Placeholder counts approximate half-precision file sizes.
Your hardware
Output dimensions
Images or videos per batch. Serving concurrency and throughput require separate measurements.
Resolution and peak VRAM
Sampling and timing
One step is one scheduler iteration. Enter timings from the same device, batch, resolution and guidance settings. Fixed overhead includes encoding, decoding and other work outside sampling.
Advanced
Performance and memory results
Memory estimate
46.4 GiB needed
Choose more memory, a smaller model, or a lighter workload.
RTX 4090 24 GiB
Vertical line = usable hardware memory
- Diffusion model (FP16)23.88 GiB
- Sampling activations16.71 GiB
- ComfyUI inference reserve0.80 GiB
- Safety margin4.97 GiB
Per-stage peak
ComfyUI moves idle models to system RAM between stages; the largest stage sets the requirement. Totals include the safety margin.
- Text encode17.92 GiB
Text encoders (FP16) 15.20 + ComfyUI inference reserve 0.80
Subtotal 16.00 + safety margin (12%) 1.92 ≈ 17.92 GiB
- SamplingPeak46.35 GiB
Diffusion model (FP16) 23.88 + Sampling activations 16.71 + ComfyUI inference reserve 0.80
Subtotal 41.39 + safety margin (12%) 4.97 ≈ 46.35 GiB
Exceeds device capacity by 17.387 GiB before safety margin
- VAE decode44.48 GiB
VAE 0.46 + VAE decode activations 38.45 + ComfyUI inference reserve 0.80
Subtotal 39.71 + safety margin (12%) 4.77 ≈ 44.48 GiB
Exceeds device capacity by 15.711 GiB before safety margin
Generation performance
- Denoising steps
- 50
- Conditioning branches per step
- 1
- Latent patch tokens per sample
- 118,800
- Estimated backbone work per step
- 13,451.842 TFLOP
- Step time at single-device BF16 peak
- 81,427.617 ms
DiT work estimates cover linear layers and visual self-attention. Enter measured step time and fixed overhead under Sampling and timing.
Compare GPU capacity
Frequently asked questions
How much VRAM does HunyuanVideo T2V 720p need?
Our default estimate at the source default resolution, batch 1, FP16 text encoders on GPU and FP16 backbone is 46.4 GiB including the safety margin. These are estimates, not measured guarantees.
Can I offload text encoders?
Yes. The CPU option removes text encoder weights from the GPU estimate, trading GPU memory for system RAM and transfer time. Each component still needs to fit on its assigned device.
Does a green result guarantee compatibility?
No. Green means the estimated memory including safety headroom fits. Backend support, precision support, kernels and runtime allocation can differ. Confirm compatibility and measure your actual workflow.
Where do these estimates come from?
Architecture and file sizes are traced to model sources. The precision table identifies measured GGUF rates, block storage and nominal bit widths. The source data for HunyuanVideo T2V 720p was last verified on 2026-11-20. Memory estimates use GiB.