Instructions to use microsoft/MAI-DS-R1-FP8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use microsoft/MAI-DS-R1-FP8 with Transformers:

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

pipe = pipeline("text-generation", model="microsoft/MAI-DS-R1-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("microsoft/MAI-DS-R1-FP8", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("microsoft/MAI-DS-R1-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 microsoft/MAI-DS-R1-FP8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "microsoft/MAI-DS-R1-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": "microsoft/MAI-DS-R1-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/microsoft/MAI-DS-R1-FP8

SGLang

How to use microsoft/MAI-DS-R1-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 "microsoft/MAI-DS-R1-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": "microsoft/MAI-DS-R1-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 "microsoft/MAI-DS-R1-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": "microsoft/MAI-DS-R1-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use microsoft/MAI-DS-R1-FP8 with Docker Model Runner:
```
docker model run hf.co/microsoft/MAI-DS-R1-FP8
```

zjngjng commited on Apr 16, 2025

Commit

e48a766

verified ·

1 Parent(s): e08d0d7

Add files using upload-large-folder tool

Browse files

Files changed (13) hide show

model-00146-of-000163.safetensors +3 -0
model-00147-of-000163.safetensors +3 -0
model-00148-of-000163.safetensors +3 -0
model-00149-of-000163.safetensors +3 -0
model-00150-of-000163.safetensors +3 -0
model-00151-of-000163.safetensors +3 -0
model-00152-of-000163.safetensors +3 -0
model-00153-of-000163.safetensors +3 -0
model-00154-of-000163.safetensors +3 -0
model-00155-of-000163.safetensors +3 -0
model-00156-of-000163.safetensors +3 -0
model-00157-of-000163.safetensors +3 -0
model-00158-of-000163.safetensors +3 -0

+version https://git-lfs.github.com/spec/v1
+oid sha256:6ebe5cee7366d1dcc44cda0085455c90b98bcc396df312dbf89169376b9c964e
+size 4302382720

+version https://git-lfs.github.com/spec/v1
+oid sha256:1e1b139e1c52ef7a058acc8997964a1ac12e8c4d1ea1685560ed9bfb5bedf1e7
+size 4302348376

+version https://git-lfs.github.com/spec/v1
+oid sha256:410902df8016a47a973698ea95a8290b3412f1d1ff93cc97a7673c33129ae9b5
+size 4302382464

+version https://git-lfs.github.com/spec/v1
+oid sha256:27415c8e213c48c1049c1d5a39bdebf970fa2169032a0cba84fca74b933c95ad
+size 4302348184

+version https://git-lfs.github.com/spec/v1
+oid sha256:703db2fc0aabac9f469ae343762897c0027d9f38988bd5a5dbbf0ed7b9f6f770
+size 4302382664

+version https://git-lfs.github.com/spec/v1
+oid sha256:8ecb7778929b2555efc95925b39b511d257aef1bf4a1ac438052a77697e5acf3
+size 4302348576

+version https://git-lfs.github.com/spec/v1
+oid sha256:d556c7401a7a414d539e3c6964d5c158a770e2f18b4f80d5f892d8348a8062b5
+size 4302382264

+version https://git-lfs.github.com/spec/v1
+oid sha256:cbe27a723728eb218dd8628ab2ce45d82e7132172931124e1c96b644332ffac2
+size 4302348392

+version https://git-lfs.github.com/spec/v1
+oid sha256:ac78863011ddef37103711f9d032526c6621a0b42527c622d3dce9f834f736fa
+size 4302382448

+version https://git-lfs.github.com/spec/v1
+oid sha256:8e53eb3ee07065be7d22e75090b8d2c549612cbb93a8f672211d5ce6e2832353
+size 4302348200

	@@ -0,0 +1,3 @@

	@@ -0,0 +1,3 @@

	@@ -0,0 +1,3 @@

	@@ -0,0 +1,3 @@