search

lmstudio-ai / lmstudio-bug-tracker

Bug tracking for the LM Studio desktop application
10 stars 3 forks source link

LM-Studio fails to import GGUF data #131

Closed sais-github closed 1 month ago

sais-github commented 1 month ago

The model on huggingface has a chat template embedded but LM-Studio is failing to import the info or even recognize it as qwen2

image image

CPU: 5600G RAM: 32GB - 28GB Usable GPU1: RTX 2080 GPU2: GTX 1660 Super GPU3: Vega-7 ((I-GPU) Not used by LM-Studio) OS: Windows 11 23H2

sais-github commented 1 month ago

It also fails to import the basic info like context length: image And the Jinja template: image

yagil commented 1 month ago

Confirmed, I see it too. Looks like a GGUF parsing issue, investigating

yagil commented 1 month ago

Fixed in 0.3.2 Build 6, available here: https://lmstudio.ai/beta-releases