vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

remove hardcoded path

Browse source 
Signed-off-by: Alex-Brooks <Alex.Brooks@ibm.com>
This commit is contained in:
Alex-Brooks 2025-01-16 15:49:05 -07:00
parent fd0111c043
commit bc66d1931b
1 changed files with 3 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

5
examples/llava/README-mmgranite.md
View file

@ -1,8 +1,9 @@
# Instructions to Convert Multimodal Granite -> GGUF
`export GRANITE_MODEL=/Users/alexanderjbrooks/Desktop/llava-granite-2b-vllm`
Disclaimer that this branch is super WIP; eventually this should be combined with the main README in this directory, but separating it for now since we use a different method for converting the LLM.
First, set the env var `$GRANITE_MODEL` to your vLLM/transformers format multimodal granite model.
`export GRANITE_MODEL=...`
### 1. Running llava surgery v2.