typed-lm-trainer v0.1.0 — LoRA/QLoRA and full fine-tuning plus FP8/FP4 post-training quantization for typed-lm dense decoder models.

Sign in to join the discussion — your thebay.events account works here.