Text Generation
Transformers
Safetensors
PyTorch
qwen2
rlhf
conversational
text-generation-inference
Instructions to use BlankZ/ragen-checkpoint-step-600-bf16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use BlankZ/ragen-checkpoint-step-600-bf16 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="BlankZ/ragen-checkpoint-step-600-bf16") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("BlankZ/ragen-checkpoint-step-600-bf16") model = AutoModelForCausalLM.from_pretrained("BlankZ/ragen-checkpoint-step-600-bf16") 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 BlankZ/ragen-checkpoint-step-600-bf16 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "BlankZ/ragen-checkpoint-step-600-bf16" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "BlankZ/ragen-checkpoint-step-600-bf16", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/BlankZ/ragen-checkpoint-step-600-bf16
- SGLang
How to use BlankZ/ragen-checkpoint-step-600-bf16 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 "BlankZ/ragen-checkpoint-step-600-bf16" \ --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": "BlankZ/ragen-checkpoint-step-600-bf16", "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 "BlankZ/ragen-checkpoint-step-600-bf16" \ --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": "BlankZ/ragen-checkpoint-step-600-bf16", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use BlankZ/ragen-checkpoint-step-600-bf16 with Docker Model Runner:
docker model run hf.co/BlankZ/ragen-checkpoint-step-600-bf16
RAGEN Checkpoint - 600 (BF16)
这是基于 Qwen/Qwen2.5-1.5B-Instruct 训练的RAGEN模型的Actor checkpoint。
模型信息
- 训练步数: 600
- 精度: BF16
- 框架: PyTorch + Transformers
- 基础模型:
Qwen/Qwen2.5-1.5B-Instruct - 任务: 文本生成 (RLHF训练后的Actor模型)
使用方法
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
model_name = "BlankZ/ragen-checkpoint-step-600-bf16"
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
# 生成文本
# 注意:请根据您的训练任务调整prompt格式
messages = [
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "你好"}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
outputs = model.generate(**model_inputs, max_length=100)
print(tokenizer.decode(outputs[0]))
注意事项
- 这是RLHF训练后的Actor模型,用于文本生成。
- 使用BF16精度以节省显存,建议使用支持BF16的GPU (如A100, H100等)。
- Downloads last month
- 2