Instructions to use lsteno/qwen2-7b-lora-opencodeinstruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lsteno/qwen2-7b-lora-opencodeinstruct with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2-7B")
model = PeftModel.from_pretrained(base_model, "lsteno/qwen2-7b-lora-opencodeinstruct")

Transformers

How to use lsteno/qwen2-7b-lora-opencodeinstruct with Transformers:

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

pipe = pipeline("text-generation", model="lsteno/qwen2-7b-lora-opencodeinstruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("lsteno/qwen2-7b-lora-opencodeinstruct", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use lsteno/qwen2-7b-lora-opencodeinstruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lsteno/qwen2-7b-lora-opencodeinstruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lsteno/qwen2-7b-lora-opencodeinstruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/lsteno/qwen2-7b-lora-opencodeinstruct

SGLang

How to use lsteno/qwen2-7b-lora-opencodeinstruct 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 "lsteno/qwen2-7b-lora-opencodeinstruct" \
    --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": "lsteno/qwen2-7b-lora-opencodeinstruct",
		"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 "lsteno/qwen2-7b-lora-opencodeinstruct" \
        --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": "lsteno/qwen2-7b-lora-opencodeinstruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use lsteno/qwen2-7b-lora-opencodeinstruct with Docker Model Runner:
```
docker model run hf.co/lsteno/qwen2-7b-lora-opencodeinstruct
```

qwen2-7b-lora-opencodeinstruct / checkpoint-50 /trainer_state.json

lsteno

Upload LoRA fine-tuned Qwen2-7B on OpenCodeInstruct

fb50571 verified 7 months ago

raw

history blame contribute delete

1.61 kB

	{
	"best_global_step": null,
	"best_metric": null,
	"best_model_checkpoint": null,
	"epoch": 2.64,
	"eval_steps": 500,
	"global_step": 50,
	"is_hyper_param_search": false,
	"is_local_process_zero": true,
	"is_world_process_zero": true,
	"log_history": [
	{
	"epoch": 0.5333333333333333,
	"grad_norm": 0.24305464327335358,
	"learning_rate": 0.00018,
	"loss": 0.4307,
	"step": 10
	},
	{
	"epoch": 1.0533333333333332,
	"grad_norm": 0.12696263194084167,
	"learning_rate": 0.00016170212765957446,
	"loss": 0.2921,
	"step": 20
	},
	{
	"epoch": 1.5866666666666667,
	"grad_norm": 0.18280404806137085,
	"learning_rate": 0.00011914893617021277,
	"loss": 0.2568,
	"step": 30
	},
	{
	"epoch": 2.1066666666666665,
	"grad_norm": 0.17879144847393036,
	"learning_rate": 7.659574468085106e-05,
	"loss": 0.2264,
	"step": 40
	},
	{
	"epoch": 2.64,
	"grad_norm": 0.2087363600730896,
	"learning_rate": 3.4042553191489365e-05,
	"loss": 0.1876,
	"step": 50
	}
	],
	"logging_steps": 10,
	"max_steps": 57,
	"num_input_tokens_seen": 0,
	"num_train_epochs": 3,
	"save_steps": 50,
	"stateful_callbacks": {
	"TrainerControl": {
	"args": {
	"should_epoch_stop": false,
	"should_evaluate": false,
	"should_log": false,
	"should_save": true,
	"should_training_stop": false
	},
	"attributes": {}
	}
	},
	"total_flos": 6.920483063935795e+16,
	"train_batch_size": 2,
	"trial_name": null,
	"trial_params": null
	}