Wan 2.1 T2V 14B on NVIDIA RTX 6000 Ada Generation

Can the NVIDIA RTX 6000 Ada Generation run Wan 2.1 T2V 14B?

Yes, but slowly

Slow. At FP8 it needs about 38.3 GB of 48.0 GB usable and takes about 27.9 min per clip.

Best precision: FP8
Memory mode: Everything on GPU
Time per clip: 27.9 min
Output: 1280×720, 81 frames, 50 steps

Best app to run it on the NVIDIA RTX 6000 Ada Generation

On Windows

ComfyUI

Desktop app · runs on PyTorch (ComfyUI), GGUF via ComfyUI-GGUF

Runs this model with built-in or popular custom nodes; GGUF files load with ComfyUI-GGUF.

Get ComfyUI

On Linux

ComfyUI

Desktop app · runs on PyTorch (ComfyUI), GGUF via ComfyUI-GGUF

Runs this model with built-in or popular custom nodes; GGUF files load with ComfyUI-GGUF.

Get ComfyUI

How we choose: the files published for this model, your operating system and GPU, whether it fits fully in memory, and published speed tests (MLX vs llama.cpp on Apple silicon, vLLM vs llama.cpp vs Ollama on NVIDIA). Details.

Download Wan 2.1 T2V 14B

Quantized files · GGUF for ComfyUI

Smaller GGUF versions for ComfyUI (with the ComfyUI-GGUF node). No sign-in needed; the model’s license still applies.

huggingface.co/city96/Wan2.1-T2V-14B-gguf (opens in a new tab)

Original weights · Hugging Face

Full-precision files from Alibaba, for fine-tuning, vLLM, Transformers, Diffusers or ComfyUI.

huggingface.co/Wan-AI/Wan2.1-T2V-14B (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from Alibaba. The weights themselves download from Hugging Face.

github.com/Wan-Video/Wan2.1 (opens in a new tab)

git clone https://github.com/Wan-Video/Wan2.1

Links checked 3 days ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 76Great quality (A) · 76 of 100: how capable the model is at FP8, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at FP8, from its editorial quality tier scaled by the quantization or precision. It’s the same on any hardware that runs this variant. Scale: A+ 85+ · A 70+ · B 55+ · C 40+ · D under 40.
Runs Slow: How well it runs on the NVIDIA RTX 6000 Ada Generation: speed, capped by how well it fits in memory. Runs great: 30+ tokens/s (or near the image, video or decision speed target) · Runs well: 10+ tokens/s · Slow: 3+ tokens/s · Very slow: under 3 · Won’t run: doesn’t fit even with offload. A tight fit or RAM offload caps it at Runs well; heavy offload caps it at Slow.

Match score

To order lists, each result also gets a match score out of 100 from four parts. It favors the most capable model that still runs well.

Model qualityEditorial quality tier of the model, scaled by the quantization or precision used.	34.4 / 45
Memory fitFull marks when everything sits in GPU memory; fewer when part spills to system RAM.	25.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.	4.9 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.	15.0 / 15
Match score	79.3 / 100

The match score orders results; it is not shown as a grade. Quality and Runs are the two ratings in the tables.

Every precision on the NVIDIA RTX 6000 Ada Generation

Quant	Memory needed	Available	Speed	Memory mode	Runs	Quality	Download
`FP16`	28.2 GB	48.0 GB	40.3 min	Sequential CPU offload	Very slowCPU offload	78Great quality (A) · 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking.	Original weights (opens in a new tab)
`FP8`	38.3 GB	48.0 GB	27.9 min	Everything on GPU	Slow	76Great quality (A) · 76 of 100: how capable the model is at FP8, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	Original weights (opens in a new tab)
`Q8 GGUF`	39.3 GB	48.0 GB	35.6 min	Everything on GPU	Slow	77Great quality (A) · 77 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q4 GGUF`	32.7 GB	48.0 GB	37.2 min	Everything on GPU	Slow	72Great quality (A) · 72 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)

Other hardware that runs it

78Great quality (A) · 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX PRO 6000 Blackwell FP16 · 24.4 min · Slow
72Great quality (A) · 72 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 5090 Q4 GGUF · 35.1 min · Slow (encoder in RAM)
76Great quality (A) · 76 of 100: how capable the model is at FP8, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 4090 FP8 · 39.0 min · Slow (tight fit · CPU offload)
78Great quality (A) · 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA H100 80GB SXM FP16 · 23.2 min · Slow
78Great quality (A) · 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA H800 80GB SXM FP16 · 23.2 min · Slow
78Great quality (A) · 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA H100 80GB PCIe FP16 · 28.4 min · Slow

Quick answers

How much memory does Wan 2.1 T2V 14B need on the NVIDIA RTX 6000 Ada Generation 48GB?

At FP8 it needs about 38.3 GB. The NVIDIA RTX 6000 Ada Generation 48GB has 48.0 GB usable.

How long does Wan 2.1 T2V 14B take on the NVIDIA RTX 6000 Ada Generation 48GB?

About 27.9 min per clip at FP8 and default settings, an estimate scaled from a reference timing by this hardware’s compute speed.

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.