Instructions to use rkumar1999/phi-tiny-moe-math-lean-sft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use rkumar1999/phi-tiny-moe-math-lean-sft with Transformers:

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

pipe = pipeline("text-generation", model="rkumar1999/phi-tiny-moe-math-lean-sft", 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("rkumar1999/phi-tiny-moe-math-lean-sft", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("rkumar1999/phi-tiny-moe-math-lean-sft", 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 rkumar1999/phi-tiny-moe-math-lean-sft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "rkumar1999/phi-tiny-moe-math-lean-sft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rkumar1999/phi-tiny-moe-math-lean-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/rkumar1999/phi-tiny-moe-math-lean-sft

SGLang

How to use rkumar1999/phi-tiny-moe-math-lean-sft 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 "rkumar1999/phi-tiny-moe-math-lean-sft" \
    --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": "rkumar1999/phi-tiny-moe-math-lean-sft",
		"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 "rkumar1999/phi-tiny-moe-math-lean-sft" \
        --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": "rkumar1999/phi-tiny-moe-math-lean-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use rkumar1999/phi-tiny-moe-math-lean-sft with Docker Model Runner:
```
docker model run hf.co/rkumar1999/phi-tiny-moe-math-lean-sft
```

rkumar1999 commited on Oct 17, 2025

Commit

482d1f0

verified ·

1 Parent(s): eb73939

End of training

Browse files

Files changed (2) hide show

README.md +4 -3
config.json +1 -1

README.md CHANGED Viewed

@@ -1,15 +1,16 @@
 ---
 base_model: microsoft/Phi-tiny-MoE-instruct
 library_name: transformers
-model_name: phi-tiny-moe-math-lean-sft
 tags:
 - generated_from_trainer
-- sft
 - trl
 licence: license
 ---
-# Model Card for phi-tiny-moe-math-lean-sft
 This model is a fine-tuned version of [microsoft/Phi-tiny-MoE-instruct](https://huggingface.co/microsoft/Phi-tiny-MoE-instruct).
 It has been trained using [TRL](https://github.com/huggingface/trl).

 ---
 base_model: microsoft/Phi-tiny-MoE-instruct
 library_name: transformers
+model_name: rkumar1999/phi-tiny-moe-math-lean-sft
 tags:
 - generated_from_trainer
 - trl
+- sft
+- alignment-handbook
 licence: license
 ---
+# Model Card for rkumar1999/phi-tiny-moe-math-lean-sft
 This model is a fine-tuned version of [microsoft/Phi-tiny-MoE-instruct](https://huggingface.co/microsoft/Phi-tiny-MoE-instruct).
 It has been trained using [TRL](https://github.com/huggingface/trl).

config.json CHANGED Viewed

@@ -37,6 +37,6 @@
   "sliding_window": 2047,
   "tie_word_embeddings": false,
   "transformers_version": "4.56.1",
-  "use_cache": false,
   "vocab_size": 32064
 }

   "sliding_window": 2047,
   "tie_word_embeddings": false,
   "transformers_version": "4.56.1",
+  "use_cache": true,
   "vocab_size": 32064
 }