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

Libraries

How to use microsoft/MediPhi with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("microsoft/MediPhi")
model = AutoModelForCausalLM.from_pretrained("microsoft/MediPhi")
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

vLLM

How to use microsoft/MediPhi with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/microsoft/MediPhi

SGLang

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

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

jpcorb20 commited on Jul 11, 2025

Commit

5808a44

verified ·

1 Parent(s): 9929a24

Update mergekit_config.yml

Browse files

Files changed (1) hide show

mergekit_config.yml +39 -39

mergekit_config.yml CHANGED Viewed

@@ -1,40 +1,40 @@
-base_model: ./Phi-3.5-mini-instruct/
-dtype: bfloat16
-merge_method: breadcrumbs_ties
-parameters:
-  int8_mask: 1.0
-  normalize: 0.0
-slices:
-- sources:
-  - layer_range: [0, 32]
-    model: ./pmc_mediphi3_5/
-    parameters:
-      density: 0.493602636300503
-      gamma: 1.6720241134317
-      weight: 0.06352674368303199
-  - layer_range: [0, 32]
-    model: ./clinical_mediphi3_5/
-    parameters:
-      density: 0.13945100550914405
-      gamma: 0.6330106659403112
-      weight: 0.5538401962393489
-  - layer_range: [0, 32]
-    model: ./medcode_mediphi3_5/
-    parameters:
-      density: 0.9265424817500185
-      gamma: 1.9263067682230453
-      weight: 0.2591472607914765
-  - layer_range: [0, 32]
-    model: ./medwiki_mediphi3_5/
-    parameters:
-      density: 0.22471877839197088
-      gamma: 0.8906459105787018
-      weight: 0.017904019341096117
-  - layer_range: [0, 32]
-    model: ./guideline_mediphi3_5/
-    parameters:
-      density: 0.766362712622725
-      gamma: 0.2721241668046967
-      weight: 1.604598536238056
-  - layer_range: [0, 32]
     model: ./Phi-3.5-mini-instruct/

+base_model: ./Phi-3.5-mini-instruct/
+dtype: bfloat16
+merge_method: breadcrumbs_ties
+parameters:
+  int8_mask: 1.0
+  normalize: 0.0
+slices:
+- sources:
+  - layer_range: [0, 32]
+    model: ./MediPhi-PubMed/
+    parameters:
+      density: 0.493602636300503
+      gamma: 1.6720241134317
+      weight: 0.06352674368303199
+  - layer_range: [0, 32]
+    model: ./MediPhi-Clinical/
+    parameters:
+      density: 0.13945100550914405
+      gamma: 0.6330106659403112
+      weight: 0.5538401962393489
+  - layer_range: [0, 32]
+    model: ./MediPhi-MedCode/
+    parameters:
+      density: 0.9265424817500185
+      gamma: 1.9263067682230453
+      weight: 0.2591472607914765
+  - layer_range: [0, 32]
+    model: ./MediPhi-MedWiki/
+    parameters:
+      density: 0.22471877839197088
+      gamma: 0.8906459105787018
+      weight: 0.017904019341096117
+  - layer_range: [0, 32]
+    model: ./MediPhi-Guidelines/
+    parameters:
+      density: 0.766362712622725
+      gamma: 0.2721241668046967
+      weight: 1.604598536238056
+  - layer_range: [0, 32]
     model: ./Phi-3.5-mini-instruct/