from huggingface_hub import snapshot_download
from vllm import LLM, SamplingParams

# Consider toggling "enforce_eager" to False if you want to load the model quicker, at the expense of tokens per second.
repo = snapshot_download(repo_id="Staticaliza/Reya-Human", allow_patterns=["*.json", "*.bin", "*.safetensors"])
llm = LLM(model=repo, dtype="auto", tensor_parallel_size=torch.cuda.device_count(), enforce_eager=True, trust_remote_code=True)

# ChatML is suggested
input = """<|im_start|>system
You are Reya.<|im_end|>
<|im_start|>user
Hi.<|im_end|>
<|im_start|>assistant
"""

params = SamplingParams(
    max_tokens=256,
    temperature=1,
    top_p=0.35,
    top_k=50,
    min_p=0.05,
    presence_penalty=0,
    frequency_penalty=0,
    repetition_penalty=1,
    stop=["<|im_end|>"],
    seed=42,
)

result = llm.generate(input, params)[0].outputs[0].text
print(result)

Staticaliza/ReyaChat

ReyaChat

vLLM: Use Instruction

Staticaliza/
ReyaChat