huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-06-10 03:12:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

connecting video to qwen2

Browse Source
This commit is contained in:
Miquel Farre 2024-11-14 10:48:18 +00:00 committed by drbh
parent 5ced960f6e
commit 05464d26bf
1 changed files with 3 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
server/text_generation_server/models/vlm_causal_lm.py
View File

@ -212,6 +212,9 @@ class VlmCausalLMBatch(FlashCausalLMBatch):
processor, image_inputs, config, image_id
)
image_id += 1
elif chunk_type == "video" and config.model_type == "qwen2_vl":
# Based on Qwen2VL's video token format
full_text += f"<video>{chunk.video}</video>"
full_text = image_text_replacement_fixup(config, full_text)