Instructions to use SOULAMA/qwen2.5-coder-ft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SOULAMA/qwen2.5-coder-ft with Transformers:

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

pipe = pipeline("text-generation", model="SOULAMA/qwen2.5-coder-ft")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("SOULAMA/qwen2.5-coder-ft")
model = AutoModelForCausalLM.from_pretrained("SOULAMA/qwen2.5-coder-ft")
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 SOULAMA/qwen2.5-coder-ft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SOULAMA/qwen2.5-coder-ft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SOULAMA/qwen2.5-coder-ft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SOULAMA/qwen2.5-coder-ft

SGLang

How to use SOULAMA/qwen2.5-coder-ft 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 "SOULAMA/qwen2.5-coder-ft" \
    --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": "SOULAMA/qwen2.5-coder-ft",
		"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 "SOULAMA/qwen2.5-coder-ft" \
        --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": "SOULAMA/qwen2.5-coder-ft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use SOULAMA/qwen2.5-coder-ft with Docker Model Runner:
```
docker model run hf.co/SOULAMA/qwen2.5-coder-ft
```

qwen2.5-coder-ft / last-checkpoint /trainer_state.json

SOULAMA

Training in progress, step 100, checkpoint

6fc7bbf verified about 2 months ago

Raw

History Blame Contribute Delete

2.83 kB

	{
	"best_global_step": 100,
	"best_metric": 0.4344141185283661,
	"best_model_checkpoint": "./qwen_coder_results/checkpoint-100",
	"epoch": 0.35555555555555557,
	"eval_steps": 25,
	"global_step": 100,
	"is_hyper_param_search": false,
	"is_local_process_zero": true,
	"is_world_process_zero": true,
	"log_history": [
	{
	"epoch": 0.0035555555555555557,
	"grad_norm": 0.9296601414680481,
	"learning_rate": 0.0,
	"loss": 0.9575080871582031,
	"step": 1
	},
	{
	"epoch": 0.08888888888888889,
	"grad_norm": 0.47108322381973267,
	"learning_rate": 0.0001994398216007982,
	"loss": 0.6088941891988119,
	"step": 25
	},
	{
	"epoch": 0.08888888888888889,
	"eval_loss": 0.5020237565040588,
	"eval_runtime": 853.513,
	"eval_samples_per_second": 0.586,
	"eval_steps_per_second": 0.293,
	"step": 25
	},
	{
	"epoch": 0.17777777777777778,
	"grad_norm": 0.43385034799575806,
	"learning_rate": 0.0001921040354671897,
	"loss": 0.46405830383300783,
	"step": 50
	},
	{
	"epoch": 0.17777777777777778,
	"eval_loss": 0.4648200273513794,
	"eval_runtime": 852.8241,
	"eval_samples_per_second": 0.586,
	"eval_steps_per_second": 0.293,
	"step": 50
	},
	{
	"epoch": 0.26666666666666666,
	"grad_norm": 0.385348379611969,
	"learning_rate": 0.00017685596144214108,
	"loss": 0.4479352951049805,
	"step": 75
	},
	{
	"epoch": 0.26666666666666666,
	"eval_loss": 0.44333434104919434,
	"eval_runtime": 855.082,
	"eval_samples_per_second": 0.585,
	"eval_steps_per_second": 0.292,
	"step": 75
	},
	{
	"epoch": 0.35555555555555557,
	"grad_norm": 0.33071649074554443,
	"learning_rate": 0.00015500550034448413,
	"loss": 0.42900867462158204,
	"step": 100
	},
	{
	"epoch": 0.35555555555555557,
	"eval_loss": 0.4344141185283661,
	"eval_runtime": 855.1679,
	"eval_samples_per_second": 0.585,
	"eval_steps_per_second": 0.292,
	"step": 100
	}
	],
	"logging_steps": 25,
	"max_steps": 282,
	"num_input_tokens_seen": 0,
	"num_train_epochs": 1,
	"save_steps": 50,
	"stateful_callbacks": {
	"EarlyStoppingCallback": {
	"args": {
	"early_stopping_patience": 5,
	"early_stopping_threshold": 0.0
	},
	"attributes": {
	"early_stopping_patience_counter": 0
	}
	},
	"TrainerControl": {
	"args": {
	"should_epoch_stop": false,
	"should_evaluate": false,
	"should_log": false,
	"should_save": true,
	"should_training_stop": false
	},
	"attributes": {}
	}
	},
	"total_flos": 1.1711404920489984e+16,
	"train_batch_size": 2,
	"trial_name": null,
	"trial_params": null
	}