change to phi3 vision
Browse files
app.py
CHANGED
|
@@ -4,14 +4,7 @@ from huggingface_hub import InferenceClient
|
|
| 4 |
"""
|
| 5 |
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
|
| 6 |
"""
|
| 7 |
-
|
| 8 |
-
"meta-llama/Meta-Llama-3-8B-Instruct",
|
| 9 |
-
"microsoft/Phi-3-vision-128k-instruct",
|
| 10 |
-
"Qwen/Qwen2-0.5B-Instruct",
|
| 11 |
-
]
|
| 12 |
-
client1 = InferenceClient(model=choices[0])
|
| 13 |
-
client2 = InferenceClient(model=choices[1])
|
| 14 |
-
client3 = InferenceClient(model=choices[2])
|
| 15 |
|
| 16 |
|
| 17 |
def respond(
|
|
@@ -21,7 +14,6 @@ def respond(
|
|
| 21 |
max_tokens,
|
| 22 |
temperature,
|
| 23 |
top_p,
|
| 24 |
-
model,
|
| 25 |
):
|
| 26 |
messages = [{"role": "system", "content": system_message}]
|
| 27 |
|
|
@@ -35,9 +27,6 @@ def respond(
|
|
| 35 |
|
| 36 |
response = ""
|
| 37 |
|
| 38 |
-
client = (
|
| 39 |
-
client1 if model == choices[0] else client2 if model == choices[1] else client3
|
| 40 |
-
)
|
| 41 |
for message in client.chat_completion(
|
| 42 |
messages,
|
| 43 |
max_tokens=max_tokens,
|
|
@@ -67,7 +56,6 @@ demo = gr.ChatInterface(
|
|
| 67 |
step=0.05,
|
| 68 |
label="Top-p (nucleus sampling)",
|
| 69 |
),
|
| 70 |
-
gr.Dropdown(choices=choices, value=choices[0], label="Model"),
|
| 71 |
],
|
| 72 |
multimodal=True,
|
| 73 |
)
|
|
|
|
| 4 |
"""
|
| 5 |
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
|
| 6 |
"""
|
| 7 |
+
client = InferenceClient(model="microsoft/Phi-3-vision-128k-instruct")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
|
| 9 |
|
| 10 |
def respond(
|
|
|
|
| 14 |
max_tokens,
|
| 15 |
temperature,
|
| 16 |
top_p,
|
|
|
|
| 17 |
):
|
| 18 |
messages = [{"role": "system", "content": system_message}]
|
| 19 |
|
|
|
|
| 27 |
|
| 28 |
response = ""
|
| 29 |
|
|
|
|
|
|
|
|
|
|
| 30 |
for message in client.chat_completion(
|
| 31 |
messages,
|
| 32 |
max_tokens=max_tokens,
|
|
|
|
| 56 |
step=0.05,
|
| 57 |
label="Top-p (nucleus sampling)",
|
| 58 |
),
|
|
|
|
| 59 |
],
|
| 60 |
multimodal=True,
|
| 61 |
)
|