# Qwen-Image-2.1 by Comfy Org: Open-Weight Model Source: https://savrn.com/models/qwen-image-2-1 Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## SAVRN's Notes on Qwen-Image-2.1 No parameter count, no context, no memory table: what the page gives you is 12 files totalling 85.6 GB of safetensors, repackaged by Comfy Org for ComfyUI from Qwen's Qwen-Image-2.1. This is the form users actually download it in — 5M pulls a month — so plan disk and VRAM from the upstream card, not from anything stated here, and budget for a workstation-class GPU. The license reads 'other', which means the obligations live in Qwen's upstream terms; read them before commercial use rather than assuming a permissive grant. The bundle includes the prompt-enhancement models for the TextGenerate node, and the card points to Qwen/Qwen-Image-2.1-PE-T2I and Qwen/Qwen-Image-2.1-PE-I2I for that purpose. Pin the revision you pull; the repository evolves with ComfyUI packaging, and 85.6 GB is a slow redownload. SAVRN Research ## Model Card Repackaged model files for ComfyUI. - https://huggingface.co/Qwen/Qwen-Image-2.1 - https://huggingface.co/Qwen/Qwen-Image-2.1-Turbo The qwen3.59bqwenimage2.1pe models are for prompt enhancement, to be used with the Generate Text node. - https://huggingface.co/Qwen/Qwen-Image-2.1-PE-T2I - https://huggingface.co/Qwen/Qwen-Image-2.1-PE-I2I For faster inference, use either the full Turbo model, or the Turbo LoRA extracted from the base model. Use 8 steps and cfg 1. Place the files in the following folders Excerpt from the card by Comfy Org, licensed other. ## Identity and Version Repository Comfy-Org/Qwen-Image-2.1 Publisher Comfy Org Task Not stated by the source Modality Other Library diffusion-single-file Parameters Not stated by the source Languages Not stated by the source Revision 0217efd1d9c9452e84f6448903d6e21bfa171ffd First published 2026-09-15 Last updated 2026-10-22 ## Files and Weights 17 files, 121.5 GB in total. The weights are 15 files totalling 121.5 GB in safetensors. Weights15 files · 121.5 GB Documentation1 file · 7.0 KB Repository1 file · 1.5 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | diffusion_models/qwen_image_2.1_bf16.safetensors | Weights | 14.2 GB | 89f4158d066c | | diffusion_models/qwen_image_2.1_int8_convrot.safetensors | Weights | 7.3 GB | cb74113cb03f | | diffusion_models/qwen_image_2.1_turbo_bf16.safetensors | Weights | 14.2 GB | 8b74237a7f50 | | diffusion_models/qwen_image_2.1_turbo_int8_convrot.safetensors | Weights | 7.3 GB | 8144a0182964 | | loras/qwen_image_2.1_turbo_lora_avg_rank_178_bf16.safetensors | Weights | 913.3 MB | 208dd43250e1 | | model_patches/qwen_image_2.1_fun_controlnet_union_bf16.safetensors | Weights | 7.6 GB | 1f9dea8384ba | | model_patches/qwen_image_2.1_fun_controlnet_union_int8_convrot.safetensors | Weights | 3.8 GB | 07aa961570ac | | text_encoders/qwen3.5_9b_qwen_image_2.1_pe_i2i.int8_convrot.safetensors | Weights | 9.5 GB | 32707d01b427 | | text_encoders/qwen3.5_9b_qwen_image_2.1_pe_i2i_w4a8.safetensors | Weights | 6.7 GB | 3c7c88f786f1 | | text_encoders/qwen3.5_9b_qwen_image_2.1_pe_t2i.int8_convrot.safetensors | Weights | 9.5 GB | 9182abae56fe | | text_encoders/qwen3.5_9b_qwen_image_2.1_pe_t2i_w4a8.safetensors | Weights | 6.7 GB | bdc244869e71 | | text_encoders/qwen3vl_8b_bf16.safetensors | Weights | 17.5 GB | 68bdc82bc1b6 | | text_encoders/qwen3vl_8b_int8_convrot.safetensors | Weights | 9.4 GB | 8bfd0f6e12ab | | text_encoders/qwen3vl_8b_w4a8.safetensors | Weights | 6.3 GB | 7754425e55e7 | | vae/qwen_image_2.1_vae_bf16.safetensors | Weights | 675.5 MB | bb21f7473051 | | README.md | Documentation | 7.0 KB | — | | .gitattributes | Repository | 1.5 KB | — | ## License and Download License other Access Open weights, no gate Download size 121.5 GB [Download from Comfy Org](https://huggingface.co/Comfy-Org/Qwen-Image-2.1) Released by Comfy Org through its official repository on Hugging Face. [Read the license](https://huggingface.co/Qwen/Qwen-Image-2.1/blob/main/LICENSE). ## Built From - Derived from Qwen/Qwen-Image-2.1 ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 121.5 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Questions About Qwen-Image-2.1 ### What license is Qwen-Image-2.1 released under? other, as its publisher declares it. Read the license text before commercial use. ## Comfy Org [All models and datasets](https://savrn.com/model-publishers/comfy-org) ## Versions - [0217efd1d9c9](https://savrn.com/models/qwen-image-2-1/versions/0217efd1d9c9) · current 2026-10-23 - [a8342359c619](https://savrn.com/models/qwen-image-2-1/versions/a8342359c619) 2026-10-22 - [df94239739ee](https://savrn.com/models/qwen-image-2-1/versions/df94239739ee) 2026-10-13 - [cb504a409072](https://savrn.com/models/qwen-image-2-1/versions/cb504a409072) 2026-10-09 - [9a44dbdb47ce](https://savrn.com/models/qwen-image-2-1/versions/9a44dbdb47ce) 2026-09-29 - [5dc5850eb514](https://savrn.com/models/qwen-image-2-1/versions/5dc5850eb514) 2026-09-23 ## Explore More - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-10-23. - [Comfy Org website](https://comfy.org/) - [Hugging Face record](https://huggingface.co/Comfy-Org/Qwen-Image-2.1) - [How the hub is built](https://savrn.com/model-hub/methodology)