Hugging Face's logo

TMElyralab
/
lyraChatGLM

English
LLM
ChatGLM6B
Model card Files Community
49
lyraChatGLM / README.md
bigmoyan's picture
bigmoyan
Update README.md
05581a1
preview code
|
raw
history blame
2.81 kB
metadata
license: creativeml-openrail-m
language:
  - en
tags:
  - LLM
  - tensorRT
  - chatGLM

Model Card for lyraChatGLM

lyraChatGLM is currently the fastest chatGLM-6B available, as far as we know, it is also the fisrt accelerated version of chatGLM-6B.

The inference speed of lyraChatGLM is 10x faster than the original version, and we're still working to improve the performance.

Among its main features are:

  • weights: original ChatGLM-6B weights released by THUDM.
  • device: lyraChatGLM is mainly based on FasterTransformer compiled for SM=80 (A100, for example).

Speed

test environment

  • device: Nvidia A100 40G
version speed
original 30 tokens/s
lyraChatGLM 310 tokens/s

Model Sources

Uses 

from transformers import AutoTokenizer
from faster_chat_glm import GLM6B, FasterChatGLM


tokenizer = AutoTokenizer.from_pretrained(chatglm6b_dir, trust_remote_code=True)

BATCH_SIZE = 8
MAX_OUT_LEN = 50

# prepare input
input_str = ["音乐推荐应该考虑哪些因素？帮我写一篇不少于800字的方案。 ", ] *
inputs = tokenizer(input_str, return_tensors="pt", padding=True)
input_ids = inputs.input_ids.to('cuda:0')


# kernel for chat model.
kernel = GLM6B(plan_path="./models/glm6b-bs{BATCH_SIZE}.ftm",
               batch_size=1,
               num_beams=1,
               use_cache=True,
               num_heads=32,
               emb_size_per_heads=128,
               decoder_layers=28,
               vocab_size=150528,
               max_seq_len=MAX_OUT_LEN)
chat = FasterChatGLM(model_dir=chatglm6b_dir, kernel=kernel).half().cuda()

# generate
sample_output = chat.generate(inputs=input_ids, max_length=MAX_OUT_LEN)
# de-tokenize model output to text
res = tokenizer.decode(sample_output[0], skip_special_tokens=True)
print(res)

Demo output

input

音乐推荐应该考虑哪些因素？帮我写一篇不少于800字的方案。

output

音乐推荐是音乐爱好者们经常面临的问题。一个好的音乐推荐应该能够根据用户的需求和喜好,推荐出符合他们口味的音乐。本文将探讨音乐

Environment 

docker pull bigmoyan/lyra_aigc:v0.1

Citation 

@Misc{lyraChatGLM2023,
  author =       {Kangjian Wu, Zhengtao Wang, Bin Wu},
  title =        {lyaraChatGLM: Accelerating chatGLM by 10x+},
  howpublished = {\url{https://huggingface.co/TMElyralab/lyraChatGLM}},
  year =         {2023}
}

Report bug