Instructions to use SenseLLM/ReflectionCoder-DS-6.7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SenseLLM/ReflectionCoder-DS-6.7B with Transformers:

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

pipe = pipeline("text-generation", model="SenseLLM/ReflectionCoder-DS-6.7B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("SenseLLM/ReflectionCoder-DS-6.7B")
model = AutoModelForCausalLM.from_pretrained("SenseLLM/ReflectionCoder-DS-6.7B")
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 SenseLLM/ReflectionCoder-DS-6.7B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SenseLLM/ReflectionCoder-DS-6.7B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SenseLLM/ReflectionCoder-DS-6.7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SenseLLM/ReflectionCoder-DS-6.7B

SGLang

How to use SenseLLM/ReflectionCoder-DS-6.7B 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 "SenseLLM/ReflectionCoder-DS-6.7B" \
    --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": "SenseLLM/ReflectionCoder-DS-6.7B",
		"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 "SenseLLM/ReflectionCoder-DS-6.7B" \
        --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": "SenseLLM/ReflectionCoder-DS-6.7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use SenseLLM/ReflectionCoder-DS-6.7B with Docker Model Runner:
```
docker model run hf.co/SenseLLM/ReflectionCoder-DS-6.7B
```

renhouxing commited on Jul 1, 2024

Commit

1fbe5b3

1 Parent(s): 7aad809

add chat template

Browse files

Files changed (2) hide show

README.md +17 -4
tokenizer_config.json +2 -1

README.md CHANGED Viewed

@@ -45,12 +45,25 @@ ReflectionCoder is a novel approach that effectively leverages reflection sequen
 Following chat templates of most models, we use two special tokens to wrap the message of user and assistant, *i.e.*, ``<|user|>``, ``<|assistant|>``, and ``<|endofmessage|>``. Furthermore, we use two special tokens to wrap the content of different blocks, *i.e.*,  ``<|text|>`` and ``<|endofblock|>``. You can use the following template to prompt our ReflectionCoder.
 ```python
-<|user|><|text|>
-Your Instruction
-<|endofblock|><|endofmessage|><|assistant|>
 ```
-#### Inference Code
 Please refer to our [GitHub Repo](https://github.com/SenseLLM/ReflectionCoder) for more technical details.
 ## Citation

 Following chat templates of most models, we use two special tokens to wrap the message of user and assistant, *i.e.*, ``<|user|>``, ``<|assistant|>``, and ``<|endofmessage|>``. Furthermore, we use two special tokens to wrap the content of different blocks, *i.e.*,  ``<|text|>`` and ``<|endofblock|>``. You can use the following template to prompt our ReflectionCoder.
 ```python
+import torch
+from transformers import pipeline
+chat = [
+    {"role": "user", "content": "<Your code instruction here>"}
+]
+generator = pipeline(
+    model="SenseLLM/ReflectionCoder-DS-6.7B",
+    task="text-generation",
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+)
+result = generator(chat, max_length=128, num_return_sequences=1)
+print(result)
 ```
 Please refer to our [GitHub Repo](https://github.com/SenseLLM/ReflectionCoder) for more technical details.
 ## Citation

tokenizer_config.json CHANGED Viewed

@@ -253,5 +253,6 @@
   "sp_model_kwargs": {},
   "tokenizer_class": "LlamaTokenizer",
   "unk_token": null,
-  "use_default_system_prompt": false
 }

   "sp_model_kwargs": {},
   "tokenizer_class": "LlamaTokenizer",
   "unk_token": null,
+  "use_default_system_prompt": false,
+  "chat_template": "{% for message in messages %}{% if message['role'] == 'user' %}{{ '<|user|>' }}{% elif message['role'] == 'system' %}{{ '<|system|>' }}{% elif message['role'] == 'assistant' %}{{ '<|assistant|>' }}{% endif %}{% if message['content'] is string %}{{ '<|text|>' + message['content'] + '<|endofblock|>' }}{% elif obj is sequence %}{% for block in message['content'] %}{% if block['type'] == 'text' %}{{ '<|text|>' }}{% elif block['type'] == 'code' %}{{ '<|code|>' }}{% elif block['type'] == 'execution' %}{{ '<|execution|>' }}{% endif %}{{ block['content'] + '<|endofblock|>' }}{% endfor %}{% endif %}{{ '<|endofmessage|>' }}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>' }}{% endif %}"
 }