squall/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

[Bugfix] Fix Token IDs Reference for MiniCPM-V When Images are Provided With No Placeholders (#8991)

Browse Source 
Signed-off-by: Alex-Brooks <Alex.Brooks@ibm.com>
This commit is contained in:
Alex Brooks 2024-10-01 03:52:44 -06:00 committed by GitHub
parent bc4eb65b54
commit 1fe0a4264a
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/model_executor/models/minicpmv.py
View File

@ -274,8 +274,8 @@ def input_processor_for_minicpmv(ctx: InputContext, llm_inputs: LLMInputs):
get_slice_image_placeholder(image_size, num_image)
prompt = llm_inputs.get("prompt")
token_ids = llm_inputs.get("prompt_token_ids")
if prompt is None:
token_ids = llm_inputs.get("prompt_token_ids")
prompt = tokenizer.decode(token_ids)
pattern = "(<image>./</image>)"