THE MODEL DIRECTORY

Every listing. A closer look.

2,841 catalog records, ordered by name. Each profile has its own link and shows where the information came from. 717 repeat listings (training checkpoints, adapters, other formats and duplicate cloud entries, each shown on its main model’s page) and listings with no published task, size, or price are left out unless you search for them. Show all 3,558

Models A–Z

Page 12 of 60 · Coverage limits

DeepSeek · Downloadable

DeepSeek-V3.1

Text generation

Generates text for conversations and writing.

Listed 685B parameters

DeepSeek · Downloadable

DeepSeek-V3.1-Base

Text generation

Generates text for conversations and writing.

Listed 685B parameters

DeepSeek · Downloadable

DeepSeek-V3.1-Terminus

Text generation

Generates text for conversations and writing.

Listed 685B parameters

DeepSeek · Downloadable

DeepSeek-V3.2

Text generation

Generates text for conversations and writing.

Listed 685B parameters

DeepSeek · Downloadable

DeepSeek-V3.2-Exp

Text generation

Generates text for conversations and writing.

Listed 685B parameters

DeepSeek · Downloadable

DeepSeek-V3.2-Exp-Base

Text generation

Generates text for conversations and writing.

Listed 685B parameters

DeepSeek · Downloadable

DeepSeek-V3.2-Speciale

Text generation

Generates text for conversations and writing.

Listed 685B parameters

DeepSeek · Downloadable

DeepSeek-V4-Flash

Text generation

Generates text for conversations and writing.

Listed 291B parameters

DeepSeek · Downloadable

DeepSeek-V4-Flash-0731

Text generation

Generates text for conversations and writing.

+1 cloud listing

Listed 304B parameters

DeepSeek · Downloadable

DeepSeek-V4-Flash-Base

Text generation

Generates text, according to its published model architecture (DeepseekV4ForCausalLM). No task is listed on Hugging Face, so check the model card for intended use.

Listed 292B parameters

DeepSeek · Downloadable

DeepSeek-V4-Flash-DSpark

Text generation

Generates text for conversations and writing.

Listed 165B parameters

DeepSeek · Downloadable

DeepSeek-V4-Pro

Text generation

Generates text for conversations and writing.

+1 cloud listing

Listed 1.6T parameters

DeepSeek · Downloadable

DeepSeek-V4-Pro-0813

Text generation

Generates text for conversations and writing.

Listed 1.7T parameters

DeepSeek · Downloadable

DeepSeek-V4-Pro-Base

Text generation

Generates text, according to its published model architecture (DeepseekV4ForCausalLM). No task is listed on Hugging Face, so check the model card for intended use.

Listed 1.6T parameters

DeepSeek · Downloadable

DeepSeek-V4-Pro-DSpark

Text generation

Generates text for conversations and writing.

Listed 1.7T parameters

DeepSeek · Downloadable

DeepSeek-V4.1-Flash

Vision

Understands text and images and responds in text.

Listed 763B parameters

DeepSeek · Downloadable

DeepSeek-V4.1-Flash-1017

Vision

Understands text and images and responds in text.

+1 cloud listing

Listed 763B parameters

DeepSeek · Downloadable

deepseek-vl-1.3b-base

Capabilities unconfirmed

A specialized research model. See the model card for its intended task and supported setup.

Listed 1.3B parameters

DeepSeek · Downloadable

deepseek-vl-1.3b-chat

Vision

Understands text and images and responds in text.

Listed 1.3B parameters

DeepSeek · Downloadable

deepseek-vl-7b-base

Capabilities unconfirmed

A specialized research model. See the model card for its intended task and supported setup.

Listed 7B parameters

DeepSeek · Downloadable

deepseek-vl-7b-chat

Vision

Understands text and images and responds in text.

Listed 7B parameters

DeepSeek · Downloadable

deepseek-vl2

Vision

Understands text and images and responds in text.

Listed 27B parameters

DeepSeek · Downloadable

deepseek-vl2-small

Vision

Understands text and images and responds in text.

Listed 16B parameters

DeepSeek · Downloadable

deepseek-vl2-tiny

Vision

Understands text and images and responds in text.

Listed 3.4B parameters

LightOn · Downloadable

DenseOn

Embeddings

Turns content into numerical representations for semantic search and retrieval.

Listed 149M parameters

LightOn · Downloadable

DenseOn-unsupervised

Embeddings

Turns content into numerical representations for semantic search and retrieval.

Listed 149M parameters

OpenBMB · Downloadable

DensingLaw-ScalingModels

Text generation

Generates text for conversations and writing.

Listed Size not disclosed

Tencent · Downloadable

DepthCrafter

VisionDepth estimation

Hugging Face lists its task as depth estimation. See the model card for supported inputs and setup.

Listed Size not disclosed

Meta · Downloadable

DepthLM

Vision

Understands text and images and responds in text.

Listed 13B parameters

Mistral AI · Downloadable

Devstral-2-123B-Instruct-2512

Text generationCoding

Generates text, according to its published model architecture (Ministral3ForCausalLM). No task is listed on Hugging Face, so check the model card for intended use.

Listed 123B parameters

Mistral AI · Downloadable

Devstral-Small-2-24B-Instruct-2512

Capabilities unconfirmed

A specialized research model. See the model card for its intended task and supported setup.

Listed 24B parameters

Mistral AI · Downloadable

Devstral-Small-2505

Text generationCoding

Generates text, according to its published model architecture (MistralForCausalLM). No task is listed on Hugging Face, so check the model card for intended use.

+1 other format

Listed 24B parameters

Mistral AI · Downloadable

Devstral-Small-2507

Text generationCoding

Generates text, according to its published model architecture (MistralForCausalLM). No task is listed on Hugging Face, so check the model card for intended use.

+1 other format

Listed 24B parameters

DeepSeek · Downloadable

dflash_gemma4_12b_block7

Capabilities unconfirmed

A specialized research model. See the model card for its intended task and supported setup.

Listed 12B parameters

DeepSeek · Downloadable

dflash_qwen3_14b_block7

Capabilities unconfirmed

A specialized research model. See the model card for its intended task and supported setup.

Listed 14B parameters

DeepSeek · Downloadable

dflash_qwen3_4b_block7

Capabilities unconfirmed

A specialized research model. See the model card for its intended task and supported setup.

Listed 4B parameters

DeepSeek · Downloadable

dflash_qwen3_8b_block7

Capabilities unconfirmed

A specialized research model. See the model card for its intended task and supported setup.

Listed 8B parameters

OpenAI · Downloadable

diffusers-cd_bedroom256_l2

Image generation

Creates or edits images. Check the model card for supported inputs and output settings.

Listed Size not disclosed

OpenAI · Downloadable

diffusers-cd_bedroom256_lpips

Image generation

Creates or edits images. Check the model card for supported inputs and output settings.

Listed Size not disclosed

OpenAI · Downloadable

diffusers-cd_cat256_l2

Image generation

Creates or edits images. Check the model card for supported inputs and output settings.

Listed Size not disclosed

OpenAI · Downloadable

diffusers-cd_cat256_lpips

Image generation

Creates or edits images. Check the model card for supported inputs and output settings.

Listed Size not disclosed

OpenAI · Downloadable

diffusers-cd_imagenet64_l2

Image generation

Creates or edits images. Check the model card for supported inputs and output settings.

Listed 296M parameters

Listing dates are source catalog dates, not verified public releases. This growing library does not yet cover every released model.