Instructions to use pankajmathur/orca_mini_v8_0_70b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use pankajmathur/orca_mini_v8_0_70b with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="pankajmathur/orca_mini_v8_0_70b")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("pankajmathur/orca_mini_v8_0_70b")
model = AutoModelForCausalLM.from_pretrained("pankajmathur/orca_mini_v8_0_70b", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use pankajmathur/orca_mini_v8_0_70b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "pankajmathur/orca_mini_v8_0_70b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "pankajmathur/orca_mini_v8_0_70b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/pankajmathur/orca_mini_v8_0_70b

SGLang

How to use pankajmathur/orca_mini_v8_0_70b with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "pankajmathur/orca_mini_v8_0_70b" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "pankajmathur/orca_mini_v8_0_70b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "pankajmathur/orca_mini_v8_0_70b" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "pankajmathur/orca_mini_v8_0_70b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use pankajmathur/orca_mini_v8_0_70b with Docker Model Runner:
```
docker model run hf.co/pankajmathur/orca_mini_v8_0_70b
```

pankajmathur commited on Dec 15, 2024

Commit

5e33d3d

verified ·

1 Parent(s): 18e1081

Update README.md

Browse files

Files changed (1) hide show

README.md +27 -2

README.md CHANGED Viewed

@@ -39,7 +39,7 @@ Hello Orca Mini, what can you do for me?<|eot_id|>
 <|start_header_id|>assistant<|end_header_id|>
 ```
-Below shows a code example on how to use this model in default full precision (bf16) format, it requires
 ```python
 import torch
@@ -59,7 +59,7 @@ outputs = pipeline(messages, max_new_tokens=128, do_sample=True, temperature=0.0
 print(outputs[0]["generated_text"][-1])
 ```
-Below shows a code example on how to use this model in 8-bit format via bitsandbytes library
 ```python
 import torch
@@ -87,6 +87,31 @@ print(outputs[0]["generated_text"][-1])
 ```
 Below shows a code example on how to do a tool use with this model and tranformer library
 Since **orca_mini_v8_0_70b** based upon LLaMA-3.3, it supports multiple tool use formats. You can see a full guide to prompt formatting [here](https://llama.meta.com/docs/model-cards-and-prompt-formats/llama3_1/).

 <|start_header_id|>assistant<|end_header_id|>
 ```
+Below shows a code example on how to use this model in default full precision (bf16) format, it requires around ~130GB VRAM
 ```python
 import torch
 print(outputs[0]["generated_text"][-1])
 ```
+Below shows a code example on how to use this model in 4-bit format via bitsandbytes library, it requires around ~39GB VRAM
 ```python
 import torch
 ```
+Below shows a code example on how to use this model in 8-bit format via bitsandbytes library, it requires around ~69GB VRAM
+```python
+import torch
+from transformers import BitsAndBytesConfig, pipeline
+model_slug = "pankajmathur/orca_mini_v8_0_70b"
+quantization_config = BitsAndBytesConfig(
+    load_in_8bit=True
+)
+pipeline = pipeline(
+    "text-generation",
+    model=model_slug,
+    model_kwargs={"quantization_config": quantization_config},
+    device_map="auto",
+)
+messages = [
+    {"role": "system", "content": "You are Orca Mini, a helpful AI assistant."},
+    {"role": "user", "content": "Hello Orca Mini, what can you do for me?"}
+]
+outputs = pipeline(messages, max_new_tokens=128, do_sample=True, temperature=0.01, top_k=100, top_p=0.95)
+print(outputs[0]["generated_text"][-1])
+```
 Below shows a code example on how to do a tool use with this model and tranformer library
 Since **orca_mini_v8_0_70b** based upon LLaMA-3.3, it supports multiple tool use formats. You can see a full guide to prompt formatting [here](https://llama.meta.com/docs/model-cards-and-prompt-formats/llama3_1/).