Qwen-Image-Edit (2511) VRAM requirements
The editing version of Qwen-Image, release 2511. Same 20B size and the same text encoder as Qwen-Image.
TypeImage
Parameters20B
Fits entirely from11 GB
8-bit or better from24 GB
01Measured on my RTX 5060 Ti 16 GB
Q4_K_M · 1024×1024
27.2s / image
qwen-image-edit-2511-Q4_K_M.gguf
Peak VRAM15.9 GB
Run range27.1–27.3 s
02The files, and how much VRAM each needs
| File | Size | Needed | Min. VRAM | Quality | Source |
|---|---|---|---|---|---|
| 16-bit | 40.9 GB | 43.9 GB | 44 GB | the original weights | Comfy-Org/Qwen-Image-Edit_ComfyUI → |
| Q8_0 | 21.8 GB | 24.8 GB | 25 GB | practically identical to the original | unsloth/Qwen-Image-Edit-2511-GGUF → |
| FP8 | 20.5 GB | 23.5 GB | 24 GB | practically identical to the original | Comfy-Org/Qwen-Image-Edit_ComfyUI → |
| Q6_K | 16.9 GB | 19.9 GB | 20 GB | very close to the original | unsloth/Qwen-Image-Edit-2511-GGUF → |
| Q5_K_M | 15.0 GB | 18.0 GB | 19 GB | close; small differences in fine detail | unsloth/Qwen-Image-Edit-2511-GGUF → |
| Q4_K_M | 13.2 GB | 16.2 GB | 17 GB | good; some loss in fine detail and text | unsloth/Qwen-Image-Edit-2511-GGUF → |
| Q3_K_M | 9.9 GB | 12.9 GB | 13 GB | noticeable loss of detail | unsloth/Qwen-Image-Edit-2511-GGUF → |
| Q2_K | 7.5 GB | 10.5 GB | 11 GB | heavy loss; a last resort | unsloth/Qwen-Image-Edit-2511-GGUF → |
“Needed” = file + 2.2 GB working memory + 0.8 GB system reserve.
Free ComfyUI workflow, tested on my RTX 5060 Ti: GGUF Q4_K_M + 4-step LoRA workflow (.json). Drag the file into ComfyUI. All workflows →
03By amount of VRAM
04Best GPU for Qwen-Image-Edit
The cheapest cards (by launch price) that run it well, and every card sorted by memory: best GPU for Qwen-Image-Edit → Planning bigger images or longer clips? Open the calculator →
05By graphics card
| GPU | VRAM | Verdict | Best file | Needed |
|---|---|---|---|---|
| Desktop graphics cards | ||||
| RTX 2060 6 GB | 6 GB | Not practical | Q2_K | 10.5 GB |
| RTX 3050 6 GB | 6 GB | Not practical | Q2_K | 10.5 GB |
| RTX 2070 Super 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 2080 Super 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 3050 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 3060 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 3060 Ti 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 3070 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 3070 Ti 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 4060 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 4060 Ti 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 5050 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 5060 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 5060 Ti 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RX 7600 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RX 9050 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RX 9060 XT 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| Arc B570 10 GB | 10 GB | Offload only | Q2_K | 10.5 GB |
| RTX 3080 10 GB | 10 GB | Offload only | Q2_K | 10.5 GB |
| RTX 2080 Ti 11 GB | 11 GB | Tight | Q2_K | 10.5 GB |
| Arc B580 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| RTX 2060 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| RTX 3060 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| RTX 3080 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| RTX 3080 Ti 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| RTX 4070 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| RTX 4070 Super 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| RTX 4070 Ti 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| RTX 5070 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| RX 7700 XT 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| RX 9070 GRE 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| Arc A770 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RTX 4060 Ti 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RTX 4070 Ti Super 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RTX 4080 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RTX 4080 Super 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RTX 5060 Ti 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RTX 5070 Ti 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RTX 5080 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RX 7600 XT 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RX 7800 XT 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RX 7900 GRE 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RX 9060 XT 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RX 9070 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RX 9070 XT 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RX 7900 XT 20 GB | 20 GB | Runs | Q6_K | 19.9 GB |
| Arc Pro B60 24 GB | 24 GB | Runs well | FP8 | 23.5 GB |
| RTX 3090 24 GB | 24 GB | Runs well | FP8 | 23.5 GB |
| RTX 3090 Ti 24 GB | 24 GB | Runs well | FP8 | 23.5 GB |
| RTX 4090 24 GB | 24 GB | Runs well | FP8 | 23.5 GB |
| RX 7900 XTX 24 GB | 24 GB | Runs well | FP8 | 23.5 GB |
| Arc Pro B70 32 GB | 32 GB | Runs well | Q8_0 | 24.8 GB |
| RTX 5090 32 GB | 32 GB | Runs well | FP8 | 23.5 GB |
| Laptop GPUs | ||||
| RTX 3050 Laptop 4 GB | 4 GB | Not practical | Q2_K | 10.5 GB |
| RTX 3050 Ti Laptop 4 GB | 4 GB | Not practical | Q2_K | 10.5 GB |
| RTX 2060 Laptop 6 GB | 6 GB | Not practical | Q2_K | 10.5 GB |
| RTX 3050 Laptop 6 GB | 6 GB | Not practical | Q2_K | 10.5 GB |
| RTX 3060 Laptop 6 GB | 6 GB | Not practical | Q2_K | 10.5 GB |
| RTX 4050 Laptop 6 GB | 6 GB | Not practical | Q2_K | 10.5 GB |
| RTX 2070 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 2070 Super Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 2080 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 2080 Super Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 3070 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 3070 Ti Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 3080 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 4060 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 4070 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 5050 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 5060 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 5070 Laptop 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RX 7600M 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RX 7600M XT 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RX 7600S 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RX 7700S 8 GB | 8 GB | Offload only | Q4_K_M | 16.2 GB |
| RTX 4080 Laptop 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| RTX 5070 Laptop 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| RTX 5070 Ti Laptop 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| RX 7800M 12 GB | 12 GB | Tight | Q2_K | 10.5 GB |
| RTX 3080 Laptop 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RTX 3080 Ti Laptop 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RTX 4090 Laptop 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RTX 5080 Laptop 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RX 7900M 16 GB | 16 GB | Tight | Q3_K_M | 12.9 GB |
| RTX 5090 Laptop 24 GB | 24 GB | Runs well | FP8 | 23.5 GB |
| Unified memory | ||||
| Radeon 8060S (Strix Halo) 96 GB | 96 GB | Runs well | 16-bit | 43.9 GB |
| Radeon 8065S (Gorgon Halo) 160 GB | 160 GB | Runs well | 16-bit | 43.9 GB |
| Apple Silicon Macs (by memory) | ||||
| Mac 16 GB | 12.7 GB | Tight | Q2_K | 10.5 GB |
| Mac 18 GB | 14.4 GB | Tight | Q3_K_M | 12.9 GB |
| Mac 24 GB | 19.6 GB | Runs | Q5_K_M | 18.0 GB |
| Mac 32 GB | 26.8 GB | Runs well | Q8_0 | 24.8 GB |
| Mac 36 GB | 30.2 GB | Runs well | Q8_0 | 24.8 GB |
| Mac 48 GB | 40.2 GB | Runs well | Q8_0 | 24.8 GB |
| Mac 64 GB | 55.7 GB | Runs well | 16-bit | 43.9 GB |
| Mac 96 GB | 85 GB | Runs well | 16-bit | 43.9 GB |
| Mac 128 GB | 115.4 GB | Runs well | 16-bit | 43.9 GB |
| Mac 192 GB | 175.4 GB | Runs well | 16-bit | 43.9 GB |
| Mac 256 GB | 236.9 GB | Runs well | 16-bit | 43.9 GB |
| Mac 512 GB | 498.1 GB | Runs well | 16-bit | 43.9 GB |
06Text encoder, VAE and other files
| File | Folder | Size | When | |
|---|---|---|---|---|
| Qwen2.5-VL 7B FP8 qwen_2.5_vl_7b_fp8_scaled.safetensors | models/text_encoders | 9.4 GB | text encoder · smaller, recommended | Download → |
| Qwen2.5-VL 7B BF16 qwen_2.5_vl_7b.safetensors | models/text_encoders | 16.6 GB | text encoder · alternative | Download → |
| Qwen-Image VAE qwen_image_vae.safetensors | models/vae | 0.3 GB | required | Download → |
The files the official ComfyUI workflows load next to the model. Sizes read from Hugging Face (2026-10-01).
07Where the files go in ComfyUI
| File | Folder | Loader node |
|---|---|---|
| Diffusion model (.safetensors: 16-bit, FP8, INT8) | ComfyUI/models/diffusion_models | Load Diffusion Model |
| GGUF file (.gguf) | ComfyUI/models/unet | Unet Loader (GGUF) - from the ComfyUI-GGUF node pack |
| Text encoder | ComfyUI/models/text_encoders | Load CLIP / DualCLIPLoader (or the GGUF versions) |
| VAE | ComfyUI/models/vae | Load VAE |
08Measured and reported results
| Label | GPU | Setup | Result | Peak VRAM | Date | Source |
|---|---|---|---|---|---|---|
| measured | RTX 5060 Ti 16 GB | qwen-image-edit-2511-Q4_K_M.gguf · 1024x1024 · 4 steps Image editing with the Lightning 4-step LoRA (Qwen-Image-Edit-2511-Lightning-4steps-V1.0-bf16), cfg 1, euler/simple, ModelSamplingAuraFlow 3.1, CFGNorm, index_timestep_zero referen | 27.19 s / image | 15.9 GB | 2026-10-11 | |
| reported | Radeon 8060S (Strix Halo) 96 GB | Qwen-Image-Edit-2511 BF16 + 4-step LoRA · ~1.6MP (dynamic) · 4 steps “"workflow": "Qwen-Image-Edit-2511-BF16-4-Step-LoRA.json", ... "duration_seconds": 112.70737218856812” kyuz0 Strix Halo toolbox benchmark; cold run incl. model load; 20-step run = 667.24 s; date from run timestamp | 112.71 s / image | — | 2026-02-13 | raw.githubusercontent.com → |
Other people's numbers are copied as published, with a link. Send yours.
09Training a LoRA for Qwen-Image-Edit
| Trainer | VRAM | Type | Settings and quote | Source |
|---|---|---|---|---|
| SimpleTuner | 24 GB | stated minimum | int8/nf4 quantised base; validation resolution must be reduced (e.g. 768x768) on 24GB “GPU VRAM : 24 G minimum (with int8/nf4 quantisation) • 40 G+ strongly recommended” | github.com → |
| ai-toolkit | 32 GB | example run | Official example config: 3-bit (uint3) base with accuracy recovery adapter, fp8 TE, cached text embeddings, low_vram, rank 16, batch 1; same for Edit-2509 “# 3bit is required for 32GB” | github.com → |
reported From each trainer's own docs, 2026-09-25. Why the figures differ →
10Every file tracked for Qwen-Image-Edit
| File | Type | Size | Repo |
|---|---|---|---|
| qwen_image_edit_2511_bf16.safetensors (2511) | BF16 | 40.86 GB | Comfy-Org/Qwen-Image-Edit_ComfyUI → |
| qwen_image_edit_2509_bf16.safetensors (2509) | BF16 | 40.86 GB | Comfy-Org/Qwen-Image-Edit_ComfyUI → |
| qwen_image_edit_2511_fp8mixed.safetensors (2511) | FP8 | 20.53 GB | Comfy-Org/Qwen-Image-Edit_ComfyUI → |
| qwen_image_edit_2509_fp8mixed.safetensors (2509) | FP8 | 20.53 GB | Comfy-Org/Qwen-Image-Edit_ComfyUI → |
| qwen_image_edit_2511_int8_convrot.safetensors (2511) | INT8 | 20.50 GB | Comfy-Org/Qwen-Image-Edit_ComfyUI → |
| qwen_image_edit_2509_fp8_e4m3fn.safetensors (2509) | FP8 | 20.43 GB | Comfy-Org/Qwen-Image-Edit_ComfyUI → |
| qwen-image-edit-2511-BF16.gguf | BF16 | 40.87 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-F16.gguf | F16 | 40.87 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-Q8_0.gguf | Q8_0 | 21.76 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-Q6_K.gguf | Q6_K | 16.85 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-Q5_1.gguf | Q5_1 | 15.39 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-Q5_K_M.gguf | Q5_K_M | 15.03 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-Q5_0.gguf | Q5_0 | 14.40 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-Q5_K_S.gguf | Q5_K_S | 14.33 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-Q4_K_M.gguf | Q4_K_M | 13.24 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-Q4_1.gguf | Q4_1 | 12.84 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-Q4_K_S.gguf | Q4_K_S | 12.41 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-Q4_0.gguf | Q4_0 | 11.85 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-Q3_K_L.gguf | Q3_K_L | 10.58 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-Q3_K_M.gguf | Q3_K_M | 9.92 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-Q3_K_S.gguf | Q3_K_S | 9.22 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
| qwen-image-edit-2511-Q2_K.gguf | Q2_K | 7.47 GB | unsloth/Qwen-Image-Edit-2511-GGUF → |
Latest Qwen-Image-Edit is 2511 (Dec 2025). Most-downloaded GGUF is unsloth (QuantStack has 2509 only). Same 20B architecture and Qwen2.5-VL-7B text encoder as Qwen-Image.