Fetching from the wire…
01
02
03
04
05
06
07
08
redhat.com
Public MindPattern findings, entities, and graph evidence that cite this source.
Findings
6
All-time hits
6
High value
0
Last seen
Connected entities
Related findings
- 2026-04-12 / SKILLSRed Hat Training Hub v0.4.0: Unsloth-Powered LoRA/QLoRA Fine-Tuning Cuts VRAM 70% and Doubles Training Speed — 8B Models on 8GB GPUsRed Hat announced Training Hub v0.4.0 (April 1) with Unsloth backend integration for production-grade LoRA and QLoRA fine-tuning. Performance gains: ~70% less VRAM than full fine-tuning, ~2x faster training than standard LoRA pipelines. QLoRA's 4-bit NF4 quantization brings 8B model fine-tuning within 8GB VRAM on consumer hardware. Best-practice starting config: r=16 with DoRA, target_modules='all-linear', rsLoRA for high ranks. The integration includes kernel-level CUDA optimizations, memory-efficient attention, gradient checkpointing, and mixed-precision training. Also ships OSFT (Orthogonal Subspace Fine-Tuning) from Hugging Face PEFT for continual learning without catastrophic forgetting.
- 2026-04-12 / VOICESKarpathy Autoresearch Hits 66K GitHub Stars — Red Hat Runs 198 Experiments Zero Intervention on OpenShift AI in 24 HoursRed Hat published an April 7 case study running Karpathy's autoresearch framework on OpenShift AI: 198 experiments, 29 kept, 2.3% validation loss improvement, zero human intervention in 24 hours. The project has grown from 21K stars at March launch to 66K+ stars and 9,600 forks. The Red Hat deployment demonstrates the pattern works at enterprise scale — an AI agent autonomously designing, running, and evaluating ML experiments on standard cloud infrastructure.
- 2026-04-02 / SKILLSTool RAG: Intelligent Tool Retrieval Triples Agent Tool Invocation Accuracy While Halving Prompt Length