mlx-community
/

Qwen2.5-7B-Instruct-4bit

Text Generation

Model card Files Files and versions Community

Update tokenizer_config.json to match Qwen/Qwen2.5-7B-Instruct

#2

by mattjcly - opened Nov 5

base: refs/heads/main

←

from: refs/pr/2

Discussion Files changed

MLX Community org Nov 5

•

There was a bug in the chat template regarding tool calls, specifically a double bracket that was fixed in https://huggingface.co./Qwen/Qwen2.5-7B-Instruct/commit/929d90a7473be218ae132181708d94dbf08a0420.

Also updates to match https://huggingface.co./Qwen/Qwen2.5-7B-Instruct/commit/acbd96531cda22292a3ceaa67e984955d3965282 in which "add_bos_token": false was removed.

Update tokenizer_config.json to match Qwen/Qwen2.5-7B-Instruct88ab1db4

prince-canuma changed pull request status to merged Nov 6

MLX Community org Nov 6

Thanks!

Merged 🚀

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment