huggingface / huggingface/candle

DeepSeek Vision Model

Open
#2,756 1 comment 3 reactions 0 assignees View on GitHub
Dominant language
Rust
Stars
21k
Forks
1.8k
Avg merge
16h 42m
Merged PRs (30d)
25

Description

Great work on Deep Seek LLM integration.
Deep Seek also has an amazing vision model DeepSeek VL
https://github.com/deepseek-ai/DeepSeek-VL?tab=readme-ov-file

there is a HF Space with it to test out.
https://huggingface.co/spaces/bobber/DeepSeek-VL-7B
lowering the repeat penalty to 1.0 gives more detailed responses. Its pretty amazing.

Contributor guide

No contributing guide indexed for this repository

Research direction

Start by reviewing Candle's existing DeepSeek LLM integration and the DeepSeek-VL repository and Hugging Face Space linked in the issue. Clarify the expected model support, input and output behavior, and validation criteria before identifying the relevant integration entry points and tests; done means DeepSeek VL works in Candle with documented verification.

Written by the indexing model from the issue text.

Assessment

Tech stack
rust
Domain
computer-vision, machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.