Text encoder
gpt_oss_ 20b_ nvfp4. safetensors
Gpt Oss 20b Nvfp4 · Text encoder for ComfyUI
- File
gpt_oss_20b_nvfp4.safetensors- Type
- Text encoder
- Size
- 13.2 GB
- Folder
ComfyUI/models/text_encoders/- Source
- Comfy-Org/Lens
- SHA-256
103d7759c720…
2 workflows use this file
What is Gpt Oss 20b Nvfp4?
Gpt Oss 20b Nvfp4 is a text encoder that converts prompts into embeddings used to guide generation. You can run it locally in ComfyUI with full control over every parameter, or access it through Comfy Cloud. ComfyUI's node-based workflow editor lets you connect Gpt Oss 20b Nvfp4 with other models and custom nodes to build any pipeline you need. There are 2 official ComfyUI workflow templates that use Gpt Oss 20b Nvfp4, ready to load and customize.
Read the full tutorial →Workflows for this file
- Z-Image-Turbo
Z-Image-Turbo Text to Image
- Flux, Flux.2 Dev
Flux.2 Dev: Image Edit
- Flux, Flux.2 Klein
Flux.2 [Klein] 9B: Image Edit
- Flux, Flux.2 Dev
Flux.2 Dev Text to Image
- Z-Image
Z-Image: Text to Image
- Anima
Anima Anime Text-to-Image Generation
- Qwen, Qwen-Image
Qwen Image 2512
- Flux, Flux.2 Klein
Flux.2 [Klein] 9B: Text to Image
Frequently Asked Questions
Gpt Oss 20b Nvfp4 is a text encoder that converts prompts into embeddings used to guide generation. You can run it locally in ComfyUI with full control over every parameter, or access it through Comfy Cloud. ComfyUI's node-based workflow editor lets you connect Gpt Oss 20b Nvfp4 with other models and custom nodes to build any pipeline you need. There are 2 official ComfyUI workflow templates that use Gpt Oss 20b Nvfp4, ready to load and customize.
Follow the step-by-step tutorial at https://docs.comfy.org/tutorials/image/lens/lens. You can also load any of the 2 official ComfyUI workflow templates that use Gpt Oss 20b Nvfp4 directly in ComfyUI.
There are 2 official ComfyUI workflow templates that use Gpt Oss 20b Nvfp4. Each template is ready to run in ComfyUI and can be customized to suit your project.
ComfyUI is free and open source. Gpt Oss 20b Nvfp4 weights are available to download from Hugging Face. You only pay for compute when running on Comfy Cloud; local inference on your own hardware is always free.