Spaces:
Running
on
Zero
Running
on
Zero
CaioXapelaum
commited on
Commit
•
df6b11b
1
Parent(s):
f93e9ec
Update app.py
Browse files
app.py
CHANGED
@@ -13,35 +13,20 @@ llm = None
|
|
13 |
llm_model = None
|
14 |
|
15 |
hf_hub_download(
|
16 |
-
repo_id="
|
17 |
-
filename="
|
18 |
local_dir = "./models"
|
19 |
)
|
20 |
|
21 |
-
hf_hub_download(
|
22 |
-
repo_id="jhofseth/Reflection-Llama-3.1-70B-GGUF",
|
23 |
-
filename="Reflection-Llama-3.1-70B-IQ3_XXS.gguf",
|
24 |
-
local_dir = "./models"
|
25 |
-
)
|
26 |
-
|
27 |
-
hf_hub_download(
|
28 |
-
repo_id="bartowski/Reflection-Llama-3.1-70B-GGUF",
|
29 |
-
filename="Reflection-Llama-3.1-70B.imatrix",
|
30 |
-
local_dir = "./random"
|
31 |
-
)
|
32 |
|
33 |
def get_messages_formatter_type(model_name):
|
34 |
-
|
35 |
-
return MessagesFormatterType.LLAMA_3
|
36 |
-
else:
|
37 |
-
raise ValueError(f"Unsupported model: {model_name}")
|
38 |
-
|
39 |
|
40 |
@spaces.GPU
|
41 |
def respond(
|
42 |
message,
|
43 |
history: list[tuple[str, str]],
|
44 |
-
model,
|
45 |
system_message,
|
46 |
max_tokens,
|
47 |
temperature,
|
@@ -108,24 +93,9 @@ def respond(
|
|
108 |
outputs += output
|
109 |
yield outputs
|
110 |
|
111 |
-
description = """<p><center>
|
112 |
-
<a href="https://huggingface.co/mattshumer/ref_70_e3" target="_blank">[Reflection Llama 3.1 70B Correct Weights]</a>
|
113 |
-
<a href="https://huggingface.co/mattshumer/Reflection-Llama-3.1-70B" target="_blank">[Old Repo]</a>
|
114 |
-
<a href="https://huggingface.co/unsloth/Reflection-Llama-3.1-70B-GGUF" target="_blank">[Reflection-Llama-3.1-70B-GGUF]</a>
|
115 |
-
|
116 |
-
</center></p>
|
117 |
-
"""
|
118 |
-
|
119 |
demo = gr.ChatInterface(
|
120 |
-
respond,
|
121 |
additional_inputs=[
|
122 |
-
gr.Dropdown([
|
123 |
-
"Reflection-Llama-3.1-70B.Q3_K_L.gguf",
|
124 |
-
"Reflection-Llama-3.1-70B-IQ3_XXS.gguf"
|
125 |
-
],
|
126 |
-
value="Reflection-Llama-3.1-70B.Q3_K_L.gguf",
|
127 |
-
label="Model"
|
128 |
-
),
|
129 |
gr.Textbox(value="You are a world-class AI system, capable of complex reasoning and reflection. Reason through the query inside <thinking> tags, and then provide your final response inside <output> tags. If you detect that you made a mistake in your reasoning at any point, correct yourself inside <reflection> tags.", label="System message"),
|
130 |
gr.Slider(minimum=1, maximum=8192, value=2048, step=1, label="Max tokens"),
|
131 |
gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
|
@@ -168,8 +138,8 @@ demo = gr.ChatInterface(
|
|
168 |
undo_btn="Undo",
|
169 |
clear_btn="Clear",
|
170 |
submit_btn="Send",
|
171 |
-
title="
|
172 |
-
description=
|
173 |
chatbot=gr.Chatbot(
|
174 |
scale=1,
|
175 |
likeable=False,
|
|
|
13 |
llm_model = None
|
14 |
|
15 |
hf_hub_download(
|
16 |
+
repo_id="TheBloke/Open_Gpt4_8x7B_v0.2-GGUF",
|
17 |
+
filename="open_gpt4_8x7b_v0.2.Q5_K_M.gguf",
|
18 |
local_dir = "./models"
|
19 |
)
|
20 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
21 |
|
22 |
def get_messages_formatter_type(model_name):
|
23 |
+
return MessagesFormatterType.ALPACA
|
|
|
|
|
|
|
|
|
24 |
|
25 |
@spaces.GPU
|
26 |
def respond(
|
27 |
message,
|
28 |
history: list[tuple[str, str]],
|
29 |
+
model="open_gpt4_8x7b_v0.2.Q5_K_M.gguf",
|
30 |
system_message,
|
31 |
max_tokens,
|
32 |
temperature,
|
|
|
93 |
outputs += output
|
94 |
yield outputs
|
95 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
96 |
demo = gr.ChatInterface(
|
97 |
+
fn=respond,
|
98 |
additional_inputs=[
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
99 |
gr.Textbox(value="You are a world-class AI system, capable of complex reasoning and reflection. Reason through the query inside <thinking> tags, and then provide your final response inside <output> tags. If you detect that you made a mistake in your reasoning at any point, correct yourself inside <reflection> tags.", label="System message"),
|
100 |
gr.Slider(minimum=1, maximum=8192, value=2048, step=1, label="Max tokens"),
|
101 |
gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
|
|
|
138 |
undo_btn="Undo",
|
139 |
clear_btn="Clear",
|
140 |
submit_btn="Send",
|
141 |
+
title="OpenGPT4",
|
142 |
+
description="Chat with *GPT-4* for free!",
|
143 |
chatbot=gr.Chatbot(
|
144 |
scale=1,
|
145 |
likeable=False,
|