Instructions to use sophiargh/MNLP_M3_mcqa_model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use sophiargh/MNLP_M3_mcqa_model with Transformers:

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

pipe = pipeline("text-generation", model="sophiargh/MNLP_M3_mcqa_model")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("sophiargh/MNLP_M3_mcqa_model")
model = AutoModelForCausalLM.from_pretrained("sophiargh/MNLP_M3_mcqa_model")
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 Settings

vLLM

How to use sophiargh/MNLP_M3_mcqa_model with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "sophiargh/MNLP_M3_mcqa_model"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sophiargh/MNLP_M3_mcqa_model",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/sophiargh/MNLP_M3_mcqa_model

SGLang

How to use sophiargh/MNLP_M3_mcqa_model 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 "sophiargh/MNLP_M3_mcqa_model" \
    --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": "sophiargh/MNLP_M3_mcqa_model",
		"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 "sophiargh/MNLP_M3_mcqa_model" \
        --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": "sophiargh/MNLP_M3_mcqa_model",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use sophiargh/MNLP_M3_mcqa_model with Docker Model Runner:
```
docker model run hf.co/sophiargh/MNLP_M3_mcqa_model
```

sophiargh commited on Jun 9, 2025

Commit

042893c

verified ·

1 Parent(s): a0ac842

End of training

Browse files

Files changed (1) hide show

README.md +9 -18

README.md CHANGED Viewed

@@ -4,8 +4,6 @@ license: apache-2.0
 base_model: Qwen/Qwen3-0.6B-Base
 tags:
 - generated_from_trainer
-metrics:
-- accuracy
 model-index:
 - name: MNLP_M3_mcqa_model_3
   results: []
@@ -18,8 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [Qwen/Qwen3-0.6B-Base](https://huggingface.co/Qwen/Qwen3-0.6B-Base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2762
-- Accuracy: 0.8986
 ## Model description
@@ -45,25 +42,19 @@ The following hyperparameters were used during training:
 - gradient_accumulation_steps: 4
 - total_train_batch_size: 8
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
-- lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.01
 - num_epochs: 4
 ### Training results
-| Training Loss | Epoch  | Step  | Validation Loss | Accuracy |
-|:-------------:|:------:|:-----:|:---------------:|:--------:|
-| 0.2684        | 0.2278 | 1000  | 0.2629          | 0.8904   |
-| 0.2572        | 0.4555 | 2000  | 0.2585          | 0.8927   |
-| 0.2475        | 0.6833 | 3000  | 0.2496          | 0.8963   |
-| 0.2508        | 0.9111 | 4000  | 0.2467          | 0.8970   |
-| 0.199         | 1.1387 | 5000  | 0.2594          | 0.8981   |
-| 0.1927        | 1.3665 | 6000  | 0.2618          | 0.8987   |
-| 0.2051        | 1.5942 | 7000  | 0.2683          | 0.8995   |
-| 0.1988        | 1.8220 | 8000  | 0.2651          | 0.8989   |
-| 0.1797        | 2.0497 | 9000  | 0.2833          | 0.9      |
-| 0.1738        | 2.2774 | 10000 | 0.2861          | 0.8995   |
-| 0.1781        | 2.5052 | 11000 | 0.2762          | 0.8986   |
 ### Framework versions

 base_model: Qwen/Qwen3-0.6B-Base
 tags:
 - generated_from_trainer
 model-index:
 - name: MNLP_M3_mcqa_model_3
   results: []
 This model is a fine-tuned version of [Qwen/Qwen3-0.6B-Base](https://huggingface.co/Qwen/Qwen3-0.6B-Base) on an unknown dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.2545
 ## Model description
 - gradient_accumulation_steps: 4
 - total_train_batch_size: 8
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.01
 - num_epochs: 4
 ### Training results
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 0.2526        | 0.2597 | 1000 | 0.2546          |
+| 0.2401        | 0.5194 | 2000 | 0.2429          |
+| 0.237         | 0.7791 | 3000 | 0.2330          |
+| 0.2227        | 1.0387 | 4000 | 0.2550          |
+| 0.1778        | 1.2984 | 5000 | 0.2545          |
 ### Framework versions