Instructions to use Thimphou/MNLP_M3_SFT_code_5percent with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Thimphou/MNLP_M3_SFT_code_5percent with Transformers:

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

pipe = pipeline("text-generation", model="Thimphou/MNLP_M3_SFT_code_5percent")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Thimphou/MNLP_M3_SFT_code_5percent")
model = AutoModelForCausalLM.from_pretrained("Thimphou/MNLP_M3_SFT_code_5percent")
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 Thimphou/MNLP_M3_SFT_code_5percent with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Thimphou/MNLP_M3_SFT_code_5percent"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Thimphou/MNLP_M3_SFT_code_5percent",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Thimphou/MNLP_M3_SFT_code_5percent

SGLang

How to use Thimphou/MNLP_M3_SFT_code_5percent 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 "Thimphou/MNLP_M3_SFT_code_5percent" \
    --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": "Thimphou/MNLP_M3_SFT_code_5percent",
		"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 "Thimphou/MNLP_M3_SFT_code_5percent" \
        --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": "Thimphou/MNLP_M3_SFT_code_5percent",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Thimphou/MNLP_M3_SFT_code_5percent with Docker Model Runner:
```
docker model run hf.co/Thimphou/MNLP_M3_SFT_code_5percent
```

Thimphou commited on Jun 6, 2025

Commit

aa48295

verified ·

1 Parent(s): e4e782b

Fine-tuned Qwen model with 5.0% code feedback (seed=42, MetaMath validation)

Browse files

Files changed (2) hide show

config.json +5 -2
model.safetensors +1 -1

config.json CHANGED Viewed

@@ -7,9 +7,12 @@
   "bos_token_id": 151643,
   "eos_token_id": 151643,
   "fine_tuning_info": {
-    "code_feedback_percentage": 0.05,
     "code_feedback_samples": 150,
-    "meta_math_samples": 3000
   },
   "head_dim": 128,
   "hidden_act": "silu",

   "bos_token_id": 151643,
   "eos_token_id": 151643,
   "fine_tuning_info": {
+    "code_feedback_percentage": 0.05263157894736842,
     "code_feedback_samples": 150,
+    "meta_math_samples": 2850,
+    "seed": 42,
+    "test_data": "MetaMath only",
+    "validation_data": "MetaMath only"
   },
   "head_dim": 128,
   "hidden_act": "silu",

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:428ce943fd08188f1bac823965e930537770abf36d4a5280126849c3d00f2df5
 size 2384234968

 version https://git-lfs.github.com/spec/v1
+oid sha256:0672118511d828a8a371daf4ef28c73dee7458e40e2f697d751a498fd644e7bf
 size 2384234968