We fine-tuned a 235B MoE model on 4 GPUs — then found we didn't need it
kgritesh · hn · 2026-10-19
A detailed engineering writeup on fine-tuning a 235B-parameter MoE model using just 4 consumer GPUs: quantization and parameter-efficient techniques push VRAM requirements down enough to run the full pipeline. The twist: after all the work, the fine-tuned model didn't actually solve their problem, so the team abandoned the route. The post documents pitfalls, hardware setups and parameter choices — valuable reading for anyone attempting low-budget fine-tuning of very large models.
More from Infra
- Liquid AI Launches RIPL: Ultra-Low-Latency Inference Engine With Sub-50ms End-to-End Responses — JosephJacks_ · 2026-10-23
- OpenJet claims Qwen3.8-27B coding runs 1.9x faster than llama.cpp on one RTX 3090 — JeremyCMorgan · 2026-10-23
- Jensen Huang lays out NVIDIA's full Vera Rubin agentic system: NVL72, MGX, CMX and RTX — rohanpaul_ai · 2026-10-23
- ONNX to Transformers.js: running models directly in your browser — JosephJacks_ · 2026-10-23
- MiniMax M3.1 Flash free for 3 days on 0G, with 1M context and 512K output — MiniMax_AI · 2026-10-23
- "We bet on Cloudflare and it paid off": R2 and D1 at billion scale — dinasaur_404 · 2026-10-23