Wan

Comfy-Org/Wan_2.2_ComfyUI_Repackaged

Comfy-Org/Wan_2.2_ComfyUI_Repackaged

Parameters

17.3B

Context

Downloads

4.7M

Likes

836

Architecture

View on HuggingFace

GPU Requirements

VRAM / RAM / disk estimates by quantization and usage scenario

QuantizationMinimumRecommendedProduction
FP1616bit
69.2 GB
RAM 103.8 · Disk 89.2
75.8 GB
RAM 113.7 · Disk 115.2
86.3 GB
RAM 129.5 · Disk 154.2
FP88bit
37.4 GB
RAM 56.0 · Disk 64.1
42.5 GB
RAM 63.8 · Disk 90.1
51.6 GB
RAM 77.3 · Disk 129.1
INT88bit
37.4 GB
RAM 56.0 · Disk 64.1
42.5 GB
RAM 63.8 · Disk 90.1
51.6 GB
RAM 77.3 · Disk 129.1
AWQ4bit
22.0 GB
RAM 33.0 · Disk 52.0
26.5 GB
RAM 39.7 · Disk 78.0
34.8 GB
RAM 52.2 · Disk 117.0
GPTQ4bit
22.2 GB
RAM 33.3 · Disk 52.2
26.7 GB
RAM 40.0 · Disk 78.2
35.0 GB
RAM 52.6 · Disk 117.2
GGUF4bit
22.6 GB
RAM 33.9 · Disk 52.5
27.1 GB
RAM 40.7 · Disk 78.5
35.5 GB
RAM 53.2 · Disk 117.5

GPU Recommendations

Best ValueBEST

1× H100 SXM

80 GB VRAM

$1,170/mo

≈ 288 video-sec/GPU-h

Effective production throughput ≈ 172.8 video-sec/GPU-h

score 0.5778

Deploy on vast ↗
Cheapest

1× RTX 5090

32 GB VRAM

$225/mo

≈ 288 video-sec/GPU-h

Effective production throughput ≈ 172.8 video-sec/GPU-h

score 0.6775

Deploy on vast ↗
Performance

8× H200 SXM

1128 GB VRAM

$20,440/mo

≈ 288 video-sec/GPU-h

Effective production throughput ≈ 172.8 video-sec/GPU-h

score 0.2197

Min Complexity

1× A100 80GB

80 GB VRAM

$876/mo

≈ 288 video-sec/GPU-h

Effective production throughput ≈ 172.8 video-sec/GPU-h

score 0.5746

Deploy on vast ↗

💬 Community — real deployment experience

0 Articles · 0 Benchmarks

View Community →

GPUs that fit (single card)

RTX 5090

32 GB VRAM

$225

/mo · vast

RTX A6000

48 GB VRAM

$226

/mo · vast

L40S

48 GB VRAM

$341

/mo · vast

RTX PRO 6000 WS

48 GB VRAM

$482

/mo · vast

RTX PRO 6000 S

48 GB VRAM

$634

/mo · vast

AmciHub

AI Model Deployment & Compute Intelligence — analyze AI model requirements, GPU performance, cloud pricing and deployment costs to find the right deployment solution.

© 2026 AmciHub. All rights reserved. Model → Requirement → GPU → Benchmark → Cloud → Cost → Recommendation