No matches. Try a model format, tool name or error.
On this page
What are the best open vision models?
Compare the image models below. Read the model's description to check what it is designed to do; this list does not measure quality or show what works in your app.
License terms and download availability are separate. Read the exact upstream license before using a model.
Answer response assembled as of . This dates the returned rows, not an upstream model update or a successful model run. Check each linked file record for current availability.
Model from Hugging Face for image text to text: google/gemma-4-31B-it · transformers
License
apache-2.0. Permissive open license: commercial use, modification, and redistribution are permitted. License details
Hosting
Availability is not confirmed. Inspect the current file record before treating it as a hosted download. 12 files listed upstream.
File evidence
No exact file name or SHA-256 is recorded in this answer row. Open the file record for current file-level evidence; unknown availability is unconfirmed.
Why listed
The catalog lists it as an image model. Recorded task: image text to text. Listed formats: safetensors. Recorded license: apache-2.0. This listing does not measure quality or prove runtime support.
Model from Hugging Face for image to video: Comfy-Org/Wan_2.2_ComfyUI_Repackaged · diffusion single file
License
apache-2.0. Permissive open license: commercial use, modification, and redistribution are permitted. License details
Hosting
Availability is not confirmed. Inspect the current file record before treating it as a hosted download. 45 files listed upstream.
File evidence
No exact file name or SHA-256 is recorded in this answer row. Open the file record for current file-level evidence; unknown availability is unconfirmed.
Why listed
The catalog lists it as an image model. Recorded task: image to video. Listed formats: safetensors. Recorded license: apache-2.0. This listing does not measure quality or prove runtime support.
Model from Hugging Face for image text to text: unsloth/Qwen3.5-9B-GGUF · transformers
License
apache-2.0. Permissive open license: commercial use, modification, and redistribution are permitted. License details
Hosting
Availability is not confirmed. Inspect the current file record before treating it as a hosted download. 28 files listed upstream.
File evidence
No exact file name or SHA-256 is recorded in this answer row. Open the file record for current file-level evidence; unknown availability is unconfirmed.
Why listed
The catalog lists it as an image model. Recorded task: image text to text. Listed formats: gguf. Recorded license: apache-2.0. This listing does not measure quality or prove runtime support.
Model from Hugging Face for image text to text: DavidAU/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-NEO-MAX-MTP-GGUF
License
apache-2.0. Permissive open license: commercial use, modification, and redistribution are permitted. License details
Hosting
Availability is not confirmed. Inspect the current file record before treating it as a hosted download. 33 files listed upstream.
File evidence
No exact file name or SHA-256 is recorded in this answer row. Open the file record for current file-level evidence; unknown availability is unconfirmed.
Why listed
The catalog lists it as an image model. Recorded task: image text to text. Listed formats: gguf. Recorded license: apache-2.0. This listing does not measure quality or prove runtime support.
Model from Hugging Face for image text to text: empero-ai/Qwythos-9B-Claude-Mythos-5-1M-GGUF · gguf
License
apache-2.0. Permissive open license: commercial use, modification, and redistribution are permitted. License details
Hosting
Availability is not confirmed. Inspect the current file record before treating it as a hosted download. 16 files listed upstream.
File evidence
No exact file name or SHA-256 is recorded in this answer row. Open the file record for current file-level evidence; unknown availability is unconfirmed.
Why listed
The catalog lists it as an image model. Recorded task: image text to text. Listed formats: gguf. Recorded license: apache-2.0. This listing does not measure quality or prove runtime support.
Model from Hugging Face for image text to text: Qwen/Qwen3.5-122B-A10B · transformers
License
apache-2.0. Permissive open license: commercial use, modification, and redistribution are permitted. License details
Hosting
Availability is not confirmed. Inspect the current file record before treating it as a hosted download. 52 files listed upstream.
File evidence
No exact file name or SHA-256 is recorded in this answer row. Open the file record for current file-level evidence; unknown availability is unconfirmed.
Why listed
The catalog lists it as an image model. Recorded task: image text to text. Listed formats: safetensors. Recorded license: apache-2.0. This listing does not measure quality or prove runtime support.
Model from Hugging Face for image to text: Salesforce/blip-image-captioning-large · transformers
License
bsd-3-clause. Permissive open license: commercial use, modification, and redistribution are permitted. License details
Hosting
Availability is not confirmed. Inspect the current file record before treating it as a hosted download. 11 files listed upstream.
File evidence
No exact file name or SHA-256 is recorded in this answer row. Open the file record for current file-level evidence; unknown availability is unconfirmed.
Why listed
The catalog lists it as an image model. Recorded task: image to text. Listed formats: safetensors. Recorded license: bsd-3-clause. This listing does not measure quality or prove runtime support.
Model from Hugging Face for image text to text: Qwen/Qwen3.8-Flash-Next · transformers
License
qwen-community-license-1.0. Commercial use is permitted with obligations: Qwen Community License 1.0: commercial use, redistribution and derivatives permitted with the copyright/permission notice; products or services with more than 100,000,000 monthly active users or US$20,000,000 monthly revenue must display the model name prominently (clause 1); a Model-as-a-Service or AI Work Assistant business must obtain a separate license from Qwen before any commercial use (clause 2). License details
Hosting
Availability is not confirmed. Inspect the current file record before treating it as a hosted download. 144 files listed upstream.
File evidence
No exact file name or SHA-256 is recorded in this answer row. Open the file record for current file-level evidence; unknown availability is unconfirmed.
Why listed
The catalog lists it as an image model. Recorded task: image text to text. Listed formats: safetensors. Recorded license: qwen-community-license-1.0. This listing does not measure quality or prove runtime support.
Model from Hugging Face for image text to text: Salesforce/blip2-opt-2.7b · transformers
License
mit. Permissive open license: commercial use, modification, and redistribution are permitted. License details
Hosting
Availability is not confirmed. Inspect the current file record before treating it as a hosted download. 18 files listed upstream.
File evidence
No exact file name or SHA-256 is recorded in this answer row. Open the file record for current file-level evidence; unknown availability is unconfirmed.
Why listed
The catalog lists it as an image model. Recorded task: image text to text. Listed formats: safetensors. Recorded license: mit. This listing does not measure quality or prove runtime support.
Availability is not confirmed. Inspect the current file record before treating it as a hosted download. 37 files listed upstream.
File evidence
No exact file name or SHA-256 is recorded in this answer row. Open the file record for current file-level evidence; unknown availability is unconfirmed.
Why listed
The catalog lists it as an image model. Recorded task: image text to text. Listed formats: safetensors. This listing does not measure quality or prove runtime support.
Model from Hugging Face for image text to text: HauhauCS/Gemma4-12B-QAT-Uncensored-HauhauCS-Balanced
License
gemma. Commercial use is permitted with obligations: Gemma terms: commercial use allowed subject to Google's prohibited-use policy. License details
Hosting
Availability is not confirmed. Inspect the current file record before treating it as a hosted download. 5 files listed upstream.
File evidence
No exact file name or SHA-256 is recorded in this answer row. Open the file record for current file-level evidence; unknown availability is unconfirmed.
Why listed
The catalog lists it as an image model. Recorded task: image text to text. Listed formats: gguf. Recorded license: gemma. This listing does not measure quality or prove runtime support.
Model from Hugging Face for zero shot image classification: laion/CLIP-convnext_base_w-laion2B-s13B-b82K-augreg · open clip
License
mit. Permissive open license: commercial use, modification, and redistribution are permitted. License details
Hosting
Availability is not confirmed. Inspect the current file record before treating it as a hosted download. 17 files listed upstream.
File evidence
No exact file name or SHA-256 is recorded in this answer row. Open the file record for current file-level evidence; unknown availability is unconfirmed.
Why listed
The catalog lists it as an image model. Recorded task: zero shot image classification. Listed formats: safetensors. Recorded license: mit. This listing does not measure quality or prove runtime support.
It depends on your task, available memory, file format and license. The records below match this guide’s filters. Check the linked file evidence before choosing: a catalog match does not prove runtime support, safety or answer quality.
An answer pack is a question-specific filter over catalog records. It is not a benchmark, safety review or guarantee that a model will run.