update for salesforce xgen

app_dialogue.py

@@ -22,20 +22,20 @@ import io
 import datasets
 
 import gradio as gr
-from transformers import AutoProcessor, TextIteratorStreamer
-from transformers import Idefics2ForConditionalGeneration
+from transformers import AutoImageProcessor, TextIteratorStreamer
+from transformers import AutoModelForVision2Seq
 
 
 DEVICE = torch.device("cuda")
 MODELS = {
-    "idefics2-8b-chatty": Idefics2ForConditionalGeneration.from_pretrained(
-        "HuggingFaceM4/idefics2-8b-chatty",
+    "xgen-mm-phi3-mini-instruct-interleave-r-v1.5": AutoModelForVision2Seq.from_pretrained(
+        "Salesforce/xgen-mm-phi3-mini-instruct-interleave-r-v1.5",
         torch_dtype=torch.bfloat16,
         _attn_implementation="flash_attention_2",
     ).to(DEVICE),
 }
 PROCESSOR = AutoProcessor.from_pretrained(
-    "HuggingFaceM4/idefics2-8b",
+    "Salesforce/xgen-mm-phi3-mini-instruct-interleave-r-v1.5",
 )
 
 SYSTEM_PROMPT = [
@@ -44,7 +44,7 @@ SYSTEM_PROMPT = [
         "content": [
             {
                 "type": "text",
-                "text": "The following is a conversation between Idefics2, a highly knowledgeable and intelligent visual AI assistant created by Hugging Face, referred to as Assistant, and a human user called User. In the following interactions, User and Assistant will converse in natural language, and Assistant will do its best to answer User’s questions. Assistant has the ability to perceive images and reason about them, but it cannot generate images. Assistant was built to be respectful, polite and inclusive. It knows a lot, and always tells the truth. When prompted with an image, it does not make up facts.",
+                "text": "The following is a conversation between xgen-mm, a highly knowledgeable and intelligent visual AI assistant created by Salesforce, referred to as Assistant, and a human user called User. In the following interactions, User and Assistant will converse in natural language, and Assistant will do its best to answer User’s questions. Assistant has the ability to perceive images and reason about them, but it cannot generate images. Assistant was built to be respectful, polite and inclusive. It knows a lot, and always tells the truth. When prompted with an image, it does not make up facts.",
             },
         ],
     },
@@ -53,7 +53,7 @@ SYSTEM_PROMPT = [
         "content": [
             {
                 "type": "text",
-                "text": "Hello, I'm Idefics2, Huggingface's latest multimodal assistant. How can I help you?",
+                "text": "Hello, I'm xgen-mm v1.5, Salesforce's latest multimodal assistant with interleave capabilities. How can I help you?",
             },
         ],
     }
@@ -414,7 +414,7 @@ top_p = gr.Slider(
 
 
 chatbot = gr.Chatbot(
-    label="Idefics2-Chatty",
+    label="xgen-mm-phi3-mini-instruct-interleave-r-v1.5",
     avatar_images=[None, BOT_AVATAR],
     height=450,
 )
@@ -424,9 +424,8 @@ with gr.Blocks(
     css=""".gradio-container .avatar-container {height: 40px width: 40px !important;} #duplicate-button {margin: auto; color: white; background: #f1a139; border-radius: 100vh; margin-top: 2px; margin-bottom: 2px;}""",
 ) as demo:
 
-    gr.Markdown("# 🐶 Hugging Face Idefics2 8B Chatty")
-    gr.Markdown("In this demo you'll be able to chat with [Idefics2-8B-chatty](https://huggingface.co/HuggingFaceM4/idefics2-8b-chatty), a variant of [Idefics2-8B](https://huggingface.co/HuggingFaceM4/idefics2-8b-chatty) further fine-tuned on chat datasets.")
-    gr.Markdown("If you want to learn more about Idefics2 and its variants, you can check our [blog post](https://huggingface.co/blog/idefics2).")
+    gr.Markdown("# 🐶 Salesforce xgen-mm-phi3-mini-instruct-interleave-r-v1.5 ")
+    gr.Markdown("In this demo you'll be able to chat with [xgen-mm-phi3-mini-instruct-interleave-r-v1.5](https://huggingface.co/Salesforce/xgen-mm-phi3-mini-instruct-interleave-r-v1.5)")
     gr.DuplicateButton(value="Duplicate Space for private use", elem_id="duplicate-button")
     # model selector should be set to `visbile=False` ultimately
     with gr.Row(elem_id="model_selector_row"):