Modeldex
Drag to turnCard front showing

DeepSeekTextPromo

DeepSeek-OCR

Vision-language model from DeepSeek, 3.3B parameters.

deepseek-ai/DeepSeek-OCR
Type
Text
Quality
Not rated yet
Parameters
3.3B (3,336,106,240)
Downloads
2.1M in the last 30 days
Likes
3.4K
Access to weights
Open download
On Hugging Face since
Oct 17, 2025
Takes in
Text, Image
Puts out
Text
Access
Open weights
License
MIT
Rarity
Promo
Set number
683/1040
Open on Hugging FaceWhere to run it 2Evolution 1/2Meet its creatureBattle Qwen3-VL-2B-Instruct

License read from the model’s Hugging Face card. Stats from DeepSeek’s official Hugging Face account, refreshed daily.

How good is it?

Public leaderboard scores for DeepSeek-OCR, refreshed daily from Epoch AI and LMArena.

No public leaderboard lists DeepSeek-OCR yet. Epoch AI and LMArena cover most chat, image, and video models from the big labs; embeddings, speech, and small open models mostly go unrated.

Where to run it

2 providers sell DeepSeek-OCR through an API. Prices are per million tokens, cheapest first.

API providers

ProviderInputOutputContext
SiliconFlow (China)Freedeepseek-ai/DeepSeek-OCRFreeFree8K
Clarifaideepseek-ai/deepseek-ocr/models/DeepSeek-OCR$0.20$0.708K

From models.dev and Hugging Face, checked daily. Resellers’ listings can lag behind their price pages, so confirm with the provider before you commit.

More from DeepSeek

DeepSeek

DeepSeek V4.1 Flash 1017

CTX1M
New
Reasoning · Vision · Tools
TextImageInput$0.15/M
TextOutput$0.60/M

October 17 update of DeepSeek V4.1 Flash with improved reasoning, coding, agentic tasks, and visual understanding

Open weightsMIT1040/1040
DeepSeek

DeepSeek-V4.1-Flash

SIZE763B
89Quality
Vision
TextImageDownloads1.3M/mo
TextLikes4.6K

Vision-language model from DeepSeek, 763B parameters.

Open weightsMIT1002/1040
DeepSeek

DeepSeek-V4-Pro-0813

SIZE1.7T
89Quality
Text
TextDownloads150K/mo
TextLikes900

Language model from DeepSeek, 1.7T parameters.

Open weightsMIT979/1040
DeepSeek

DeepSeek V4 Pro

CTX1M
78Quality
Reasoning · Tools
TextInput$0.66/M
TextOutput$1.98/M

DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes

Open weightsMIT977/1040
DeepSeek

DeepSeek-V4-Flash-0731

SIZE304B
87Quality
Text
TextDownloads4.2M/mo
TextLikes4.1K

Language model from DeepSeek, 304B parameters.

Open weightsMIT959/1040
DeepSeek

DeepSeek-V4-Flash-DSpark

SIZE165B
Text
TextDownloads150K/mo
TextLikes300

Language model from DeepSeek, 165B parameters.

Open weightsMIT932/1040
DeepSeek

DeepSeek-OCR-2

SIZE3.4B
Vision
TextImageDownloads600K/mo
TextLikes1.1K

Vision-language model from DeepSeek, 3.4B parameters.

Open weightsApache 2.0762/1040
DeepSeek

DeepSeek-V3.2

SIZE685B
71Quality
Text
TextDownloads5.2M/mo
TextLikes1.6K

Language model from DeepSeek, 685B parameters.

Open weightsMIT711/1040
Deck0/4
  1. Empty slot
  2. Empty slot
  3. Empty slot
  4. Empty slot