Unsloth AI

Unsloth AI unsloth unslothaiunsloth 公开仓库 更新于1天前 Finetune Llama , Mistral, Phi & Gemma LLMs 2-5x faster with 80% less memory 13k Python THUDM963627chatglm3 pypi unsloth In this tutorial, I've shared an exciting method to speed up your large language model fine-tuning process using Unsloth

pypi unsloth unsloth lianshufengllm:unsloth Manifest digest sha256:7b7c0bf4ea4f4b92e9ae0925264c1443e981798af37db9f896100bf6d2385b2f OSARCH linuxamd64 Compressed

unsloth Blog ; Llama 3 Meta Unsloth · Mistral NeMo Bug Fixes & Ollama supportJul 18, 2024 ; Llama 3 Meta Unsloth Gemma 2 Unsloth appears to train 24% faster on a 4090 and 28% on a 3090 than torchtune with () It also uses significantly less memory allowing you to