Seeking Local Model Fine-Tuning Alternatives
DanielSReichenbach · reddit · 2026-07-14
The author currently runs models locally via pi and recently discovered that unsloth supports fine-tuning/training with custom datasets. They are looking for alternative solutions that do not depend on unsloth.
Their goal is to perform custom training for their specific workflows based on an existing llama.cpp server and a few local models, and they are seeking recommendations from the community.
More from Models
- Moonshot’s Kimi K3 reaches #5 on MathArena as the top open model — xeophon · 2026-07-22
- Google launches Gemini 3.5 Flash Cyber for CodeMender, with limited access for governments — GoogleAI · 2026-07-22
- Kimi K3 tops Gemini 3.6 Flash on four shared public benchmarks — ChrisGPT · 2026-07-22
- Google’s year-long pause in new base-model pretraining draws sharp criticism — teortaxesTex · 2026-07-22
- Current setup is 8,192 input tokens and 2,048 output tokens, with 8k/512 next — TheZachMueller · 2026-07-22
- Kimi K3 feels slower than K2.7, but stronger on long coding jobs and refactoring — Far-Presence2711 · 2026-07-22