Ver categorias
Explorar
Fiverr Pro
Português
$
USD
Bryan Ploh
@aimodelsmith
AI Engineer LLM Fine Tuning with LoRA and QLoRA
Singapura
Inglês
Sobre mim
I build custom AI models. Not prompt templates - real, fine-tuned LLMs trained on your data, tested before delivery, shipped in the format you need (GGUF/llama.cpp or HuggingFace/vLLM). 7B-14B via LoRA/QLoRA. Every model ships with a scored accuracy report on held-out questions: 80%+ or I retrain free. Trained on a private 2x32 GB GPU rig - your data never hits a public API. 7B delivers in 24-48h. Send me 100-5,000 Q&A pairs.
Habilidades
Bryan Ploh
offline •
Conheça meus serviços
Ajustes de modelo de IA
I will fine tune a 7b or 14b ai model on your data and verify it
