Hugging Face's logo

nvidia
/
llama-nemoretriever-colembed-3b-v1

Visual Document Retrieval
Transformers
Safetensors
ColPali
multilingual
llama_nemoretrievercolembed
feature-extraction
text
image
vidore
multimodal-embedding
multilingual-embedding
Text-to-Visual Document (T→VD) retrieval
custom_code
Model card Files
xet
Community
8

Enables MTEB 2 evaluation for Vidore V1-V2-V3 benchmarks

#7
by gmoreira-nv - opened
base: refs/heads/main
←
from: refs/pr/7
Discussion Files changed
+139
-35
Making ColEmbed model compatible with MTEB 2, for Vidore evaluation8aeec567
gmoreira-nv
 NVIDIA org
•
edited 15 days ago

Besides the model changes in this MR, evaluating the model on MTEB 2 depends on an MTEB MR (already merged)

 Fixed readmeea98ca2f
gmoreira-nv changed pull request title from Enable Vidore V1-V2-V3 evaluation with MTEB 2 to Enables MTEB 2 evaluation for Vidore V1-V2-V3 benchmarks
Ädding an example script to eval the model with MTEB 2 on Vidore benchmarksa18d2a9b
 Fixing readme and scriptf1d2d4b7
nvidia-oliver-holworthy
 NVIDIA org

this looks good to merge

gmoreira-nv changed pull request status to open
gmoreira-nv changed pull request status to merged

Sign up or log in to comment