Kandinsky 6.0 Pro (29B, video + audio) on NVIDIA H100 NVL

Can the NVIDIA H100 NVL run Kandinsky 6.0 Pro (29B, video + audio)?

Yes, but slowly (tight fit)

Slow (tight fit). At FP16 it needs about 85.4 GB of 94.0 GB usable and takes about 37.0 min per clip.

Best precision: FP16
Memory mode: Everything on GPU
Time per clip: 37.0 min
Output: 864×480, 121 frames, 50 steps

Best app to run it on the NVIDIA H100 NVL

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

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

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 Kandinsky 6.0 Pro (29B, video + audio)

Original weights · Hugging Face

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

huggingface.co/kandinskylab/Kandinsky-6.0-Pro-5s-Diffusers (opens in a new tab)

Free Hugging Face account needed: accept the license on the model page and the download starts right away.

Official repo · GitHub

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

github.com/kandinskylab/kandinsky-6 (opens in a new tab)

git clone https://github.com/kandinskylab/kandinsky-6

Links checked 6 days ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 88Excellent quality (A+) · 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at FP16, 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 Slowtight fit: How well it runs on the NVIDIA H100 NVL: 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.	39.6 / 45
Memory fitFull marks when everything sits in GPU memory; fewer when part spills to system RAM.	23.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.	4.3 / 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	81.9 / 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 H100 NVL

Quant	Memory needed	Available	Speed	Memory mode	Runs	Quality	Download
`FP16`	85.4 GB	94.0 GB	37.0 min	Everything on GPU	Slowtight fit	88Excellent quality (A+) · 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	Original weights (opens in a new tab)

Other hardware that runs it

88Excellent quality (A+) · 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX PRO 6000 Blackwell FP16 · 35.9 min · Slow (tight fit)
88Excellent quality (A+) · 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA H200 141GB SXM FP16 · 34.1 min · Slow
88Excellent quality (A+) · 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA H200 NVL FP16 · 37.0 min · Slow
88Excellent quality (A+) · 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GH200 Grace Hopper (96 GB GPU) FP16 · 34.1 min · Slow (tight fit)
88Excellent quality (A+) · 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX PRO 6000 Blackwell Server Edition FP16 · 37.0 min · Slow (tight fit)
88Excellent quality (A+) · 88 of 100: how capable the model is at FP16, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition FP16 · 39.5 min · Slow (tight fit)

Quick answers

How much memory does Kandinsky 6.0 Pro (29B, video + audio) need on the NVIDIA H100 NVL 94GB?

At FP16 it needs about 85.4 GB. The NVIDIA H100 NVL 94GB has 94.0 GB usable.

How long does Kandinsky 6.0 Pro (29B, video + audio) take on the NVIDIA H100 NVL 94GB?

About 37.0 min per clip at FP16 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.