Update app.py
Browse files
app.py
CHANGED
@@ -12,12 +12,12 @@ from huggingface_hub import hf_hub_download
|
|
12 |
hf_hub_download(
|
13 |
repo_id="PartAI/Dorna-Llama3-8B-Instruct-GGUF",
|
14 |
filename="dorna-llama3-8b-instruct.Q2_K.gguf",
|
15 |
-
local_dir = "
|
16 |
)
|
17 |
hf_hub_download(
|
18 |
repo_id="PartAI/Dorna-Llama3-8B-Instruct-GGUF",
|
19 |
filename="dorna-llama3-8b-instruct.Q4_0.gguf",
|
20 |
-
local_dir = "
|
21 |
)
|
22 |
|
23 |
css = """
|
@@ -57,7 +57,7 @@ def respond(
|
|
57 |
chat_template = get_messages_formatter_type(model)
|
58 |
|
59 |
llm = Llama(
|
60 |
-
model_path=f"
|
61 |
flash_attn=True,
|
62 |
n_threads=40,
|
63 |
n_gpu_layers=81,
|
|
|
12 |
hf_hub_download(
|
13 |
repo_id="PartAI/Dorna-Llama3-8B-Instruct-GGUF",
|
14 |
filename="dorna-llama3-8b-instruct.Q2_K.gguf",
|
15 |
+
local_dir = "."
|
16 |
)
|
17 |
hf_hub_download(
|
18 |
repo_id="PartAI/Dorna-Llama3-8B-Instruct-GGUF",
|
19 |
filename="dorna-llama3-8b-instruct.Q4_0.gguf",
|
20 |
+
local_dir = "."
|
21 |
)
|
22 |
|
23 |
css = """
|
|
|
57 |
chat_template = get_messages_formatter_type(model)
|
58 |
|
59 |
llm = Llama(
|
60 |
+
model_path=f"./{model}",
|
61 |
flash_attn=True,
|
62 |
n_threads=40,
|
63 |
n_gpu_layers=81,
|