Instructions to use RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8 with Transformers:

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

pipe = pipeline("text-generation", model="RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8", trust_remote_code=True)
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8

SGLang

How to use RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8 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 "RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8" \
    --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": "RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8",
		"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 "RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8" \
        --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": "RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8 with Docker Model Runner:
```
docker model run hf.co/RedHatAI/DeepSeek-Coder-V2-Lite-Instruct-FP8
```

Lin-K76 commited on Jul 18, 2024

Commit

4862bb4

verified ·

1 Parent(s): d3479a2

Update README.md

Browse files

Files changed (1) hide show

README.md +5 -2

README.md CHANGED Viewed

@@ -2,6 +2,9 @@
 tags:
 - fp8
 - vllm
 ---
 # DeepSeek-Coder-V2-Lite-Instruct-FP8
@@ -19,13 +22,13 @@ tags:
 - **Version:** 1.0
 - **Model Developers:** Neural Magic
-Quantized version of [DeepSeek-Coder-V2-Lite-Instruct](deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct).
 <!-- It achieves an average score of 73.19 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark (version 1), whereas the unquantized model achieves 73.48. -->
 It achieves an average score of 79.60 on the [HumanEval+](https://github.com/openai/human-eval?tab=readme-ov-file) benchmark, whereas the unquantized model achieves 79.33.
 ### Model Optimizations
-This model was obtained by quantizing the weights and activations of [DeepSeek-Coder-V2-Lite-Instruct](deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct) to FP8 data type, ready for inference with vLLM >= 0.5.2.
 This optimization reduces the number of bits per parameter from 16 to 8, reducing the disk size and GPU memory requirements by approximately 50%.
 Only the weights and activations of the linear operators within transformers blocks are quantized. Symmetric per-tensor quantization is applied, in which a linear scaling per output dimension maps the FP8 representations of the quantized weights and activations.

 tags:
 - fp8
 - vllm
+license: other
+license_name: deepseek-license
+license_link: LICENSE
 ---
 # DeepSeek-Coder-V2-Lite-Instruct-FP8
 - **Version:** 1.0
 - **Model Developers:** Neural Magic
+Quantized version of [DeepSeek-Coder-V2-Lite-Instruct](https://huggingface.co/deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct).
 <!-- It achieves an average score of 73.19 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark (version 1), whereas the unquantized model achieves 73.48. -->
 It achieves an average score of 79.60 on the [HumanEval+](https://github.com/openai/human-eval?tab=readme-ov-file) benchmark, whereas the unquantized model achieves 79.33.
 ### Model Optimizations
+This model was obtained by quantizing the weights and activations of [DeepSeek-Coder-V2-Lite-Instruct](https://huggingface.co/deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct) to FP8 data type, ready for inference with vLLM >= 0.5.2.
 This optimization reduces the number of bits per parameter from 16 to 8, reducing the disk size and GPU memory requirements by approximately 50%.
 Only the weights and activations of the linear operators within transformers blocks are quantized. Symmetric per-tensor quantization is applied, in which a linear scaling per output dimension maps the FP8 representations of the quantized weights and activations.