close
Jump to content

Vicuna LLM

From Wikipedia, the free encyclopedia
Vicuna
DeveloperLMSYS
ReleaseMarch 30, 2023; 3 years ago (2023-03-30)[1]
Stable release
v1.5 / August 1, 2023; 3 years ago (2023-08-01)[2]
TypeLarge language model
LicenseCode: Apache License 2.0; v1.5 weights: Llama 2 Community License[3]
Repositorygithub.com/lm-sys/FastChat

Vicuna LLM is a large language model (LLM) released by LMSYS.[1] The first versions were fine-tuned from Meta's LLaMA. Version 1.5 was fine-tuned from Llama 2.[2][3] Its training and serving code is open source,[1][4] and the model has been used in AI research.[5][6] LMSYS also ran a public online demo of the model for non-commercial use.[1]

The methodology of LMSYS's Chatbot Arena is to enable the public at large to contrast and compare the accuracy of LLMs "in the wild" (an example of citizen science) and to vote on their output; a question-and-answer chat format is used. At the beginning of each round two LLM chatbots from a diverse pool of nine are presented randomly and anonymously, their identities only being revealed upon voting on their answers. The user has the option of either replaying ("regenerating") a round, or beginning an entirely fresh one with new LLMs.[7] (The user also has the option of choosing which LLMs to do battle.)[relevant?]

References

[edit]
  1. 1 2 3 4 "Vicuna: An Open-Source Chatbot Impressing GPT-4 with 90%* ChatGPT Quality | LMSYS Org". lmsys.org. 30 March 2023.
  2. 1 2 "Vicuna Weights". GitHub. LMSYS. Retrieved 2026-10-27.
  3. 1 2 "lmsys/vicuna-13b-v1.5 · Hugging Face". huggingface.co. 18 July 2023.
  4. ↑ "FastChat". GitHub. LMSYS. Retrieved 2026-10-27.
  5. ↑ Peng, Baolin; Li, Chunyuan; He, Pengcheng; Galley, Michel; Gao, Jianfeng (2023). "Instruction Tuning with GPT-4". arXiv:2304.03277 [cs.CL].
  6. ↑ Zheng, Lianmin; Chiang, Wei-Lin; Sheng, Ying; Zhuang, Siyuan; Wu, Zhanghao; Zhuang, Yonghao; Lin, Zi; Li, Zhuohan; Li, Dacheng; Xing, Eric P.; Zhang, Hao; Gonzalez, Joseph E.; Stoica, Ion (2023). "Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena". arXiv:2306.05685 [cs.CL].
  7. ↑ "Vicuna LLM Commercially Available, New v1.5 Update Improves Context Length".
[edit]