Image-Text-to-Text
Transformers
Safetensors
English
idefics2
pretraining
multimodal
vision
Inference Endpoints
5 papers
VictorSanh commited on
Commit
45e8045
1 Parent(s): f203285

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -41,7 +41,7 @@ Idefics2 is an open multimodal model that accepts arbitrary sequences of image a
41
  We release under the Apache 2.0 license 2 checkpoints:
42
  - [idefics2-8b-base](https://huggingface.co/HuggingFaceM4/idefics2-8b-base): the base model
43
  - [idefics2-8b](https://huggingface.co/HuggingFaceM4/idefics2-8b): the base model fine-tuned on a mixture of supervised and instruction datasets (text-only and multimodal datasets)
44
- - [idefics2-8b-chatty](https://huggingface.co/HuggingFaceM4/idefics2-8b-chatty): `idefics2-8b` further fine-tuned on long conservation
45
 
46
  # Model Summary
47
 
 
41
  We release under the Apache 2.0 license 2 checkpoints:
42
  - [idefics2-8b-base](https://huggingface.co/HuggingFaceM4/idefics2-8b-base): the base model
43
  - [idefics2-8b](https://huggingface.co/HuggingFaceM4/idefics2-8b): the base model fine-tuned on a mixture of supervised and instruction datasets (text-only and multimodal datasets)
44
+ - [idefics2-8b-chatty](https://huggingface.co/HuggingFaceM4/idefics2-8b-chatty): `idefics2-8b` further fine-tuned on long conversation
45
 
46
  # Model Summary
47