Vicuna
LMSYS · March 2023
● activeOpen Sourcedecoder onlytext
Parameters13B
Why It Matters
Helped establish the LMSYS Chatbot Arena, which became the gold standard for evaluating and comparing language models through blind human preference voting.
Description
LMSYS's instruction-tuned LLaMA model trained on ~70K conversations shared by ChatGPT users on ShareGPT. Evaluated by GPT-4 as achieving over 90% of ChatGPT's quality, it became one of the most popular open-source chat models and helped establish the LMSys Chatbot Arena evaluation platform.
Key Innovations
Instruction Tuning
Instruction TuningFine-tuning a model on instruction-response pairs so it follows user commands more reliably.
More from Community / Uncensored
WizardLM2023-06 · 13B
Dolphin (Eric Hartford)2023-07 · —
Hermes (Nous Research)2023-09 · —
LLaMA 4 Scout Abliterated2025-05 · 81B (17B active × 16 experts)
DeepSeek R1 Uncensored2025-03 · 671B (37B active)
Bark2023-04 · —
Mistral 7B Uncensored2024-01 · 7B
Qwen 2.5 72B Abliterated2025-01 · 72B
SOLAR 10.7B Uncensored2024-02 · 10.7B
LLaMA 3.1 405B Abliterated2024-08 · 405B