Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -14,24 +14,17 @@ MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
|
|
14 |
DESCRIPTION = """\
|
15 |
# Labradorite-13b Chat
|
16 |
|
17 |
-
This Space demonstrates model [
|
18 |
|
19 |
"""
|
20 |
|
21 |
-
LICENSE = """
|
22 |
-
<p/>
|
23 |
-
|
24 |
-
---
|
25 |
-
As a derivate work of Labradorite-13b](https://huggingface.co/ibm/labradorite-13b) by IBM,
|
26 |
-
this demo is governed by the original [license](https://huggingface.co/spaces/Artples/Labradorite-13b-Chat/blob/main/LICENSE.txt) and [acceptable use policy](https://huggingface.co/spaces/Artples/Labradorite-13b-Chat/blob/main/USE_POLICY.md).
|
27 |
-
"""
|
28 |
|
29 |
if not torch.cuda.is_available():
|
30 |
DESCRIPTION += "\n<p>Running on CPU! This demo does not work on CPU.</p>"
|
31 |
|
32 |
|
33 |
if torch.cuda.is_available():
|
34 |
-
model_id = "ibm/
|
35 |
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", load_in_4bit=True)
|
36 |
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
37 |
tokenizer.use_default_system_prompt = False
|
|
|
14 |
DESCRIPTION = """\
|
15 |
# Labradorite-13b Chat
|
16 |
|
17 |
+
This Space demonstrates model [Merlinite-7b](https://huggingface.co/ibm/merlinite-7b) by IBM, a finetuned Mistral model with 7B parameters for chat instructions.
|
18 |
|
19 |
"""
|
20 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
21 |
|
22 |
if not torch.cuda.is_available():
|
23 |
DESCRIPTION += "\n<p>Running on CPU! This demo does not work on CPU.</p>"
|
24 |
|
25 |
|
26 |
if torch.cuda.is_available():
|
27 |
+
model_id = "ibm/merlinite-7b"
|
28 |
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", load_in_4bit=True)
|
29 |
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
30 |
tokenizer.use_default_system_prompt = False
|