Havuz

Halo is an open-source framework built by White Circle for training large language and multimodal models

Single-GPU LLM fine-tuning preflight: memory estimates, runnable LoRA/QLoRA recipes, and local measurements.

verl for a single consumer GPU. PPO, GRPO and on-policy distillation on NVIDIA GPUs.

Automated alignment adjustment for LLMs — direct steering, LoRA, and MoE expert-granular abliteration, optimized via multi-objective Optuna

Speedrunning LoRA fine-tuning: frozen task, frozen hardware, public wall-clock leaderboard. modded-nanogpt for fine-tuning.

MCore-Bridge: Providing Megatron-Core model definitions for state-of-the-art large models and making Megatron training as simple as Transfor

Distill teacher chains-of-thought into a LoRA adapter via a strict boxed-answer format contract + two-phase Train→Nudge (silver-medal NVIDIA

🚀 Engram-PEFT: An unofficial implementation of DeepSeek Engram. Inject high-capacity conditional memory into LLMs via sparse retrieval PEFT

Parameter-Efficient Fine-Tuning For Edge-Cloud Collabrative Computing

One GPU. Full LLM workflow. Real benchmarks. No cloud required.

Asking for What Was Never Requested: horizontal and vertical proactivity in LLM agents. Need-graph metrics (no LLM judge) and Q&D, which tra

A user's facts become a few rows in a content-addressed memory table — not a rewrite of the model. Paper, code, data, and interactive site f

Companion code for the Manning book 'LLM Customization and Fine-Tuning.' Adapt open-weights LLMs end to end: prompting and RAG, LoRA/QLoRA,

Sydney: Fine-tuned model, companion novel, and philosophical inquiries.

Training-data memorization auditor for fine-tuned LLMs — Trainer/TRL plugin, canary MIA + regurgitation audit, Apache-2.0