Updates and recent posts about Unsloth..

Posts
Description

Story Trending

@laura_garcia shared a post, 1 week, 3 days ago

Software Developer, RELIANOID

𝗗𝗮𝘁𝗮 + 𝗔𝗜 𝗦𝘂𝗺𝗺𝗶𝘁 𝟮𝟬𝟮𝟲

🤖 𝗗𝗮𝘁𝗮 + 𝗔𝗜 𝗦𝘂𝗺𝗺𝗶𝘁 𝟮𝟬𝟮𝟲 𝗶𝘀 𝘂𝗻𝗱𝗲𝗿𝘄𝗮𝘆! 𝗥𝗘𝗟𝗜𝗔𝗡𝗢𝗜𝗗 is following the latest innovations from Databricks Data + AI Summit 2026, where data engineers, AI practitioners, architects, and technology leaders come together to explore the future of AI, analytics, machine learning, and modern data platforms. Ex..

databricks_data_ai_summit_san_francisco_2026_june_relianoid

Story Trending

@laura_garcia shared a post, 1 week, 4 days ago

Software Developer, RELIANOID

Identity Week Europe 2026

🔐 Identity Week Europe 2026 is here! RELIANOID is attending Identity Week Europe 2026 in Amsterdam on June 9–10, bringing together the global identity ecosystem to explore the latest innovations in digital identity, IAM, authentication, biometrics, fraud prevention, and digital trust. We're looking ..

Story Trending

@laura_garcia shared a post, 1 week, 5 days ago

Software Developer, RELIANOID

RELIANOID Enterprise Edition v8.6

🚀 Introducing RELIANOID Enterprise Edition v8.6 Security hardening, long-term stability, advanced observability, and stronger application delivery — built for mission-critical infrastructures. What’s new in v8.6? 🔐 Built on Debian Bookworm 12.14 with our new long-term support kernel (6.1.174-1-noid0..

Story Trending

@laura_garcia shared a post, 1 week, 5 days ago

Software Developer, RELIANOID

RELIANOID Community Edition v7.10 is here

🚀 RELIANOID Community Edition v7.10 is here! More observability, stronger foundations, and smarter load balancing capabilities — designed to make your infrastructure more reliable and easier to operate. Highlights of v7.10: 🔐 Updated to Debian Bookworm 12.14 for improved security and stability 🌐 New..

Link

@anjali5 shared a link, 1 week, 5 days ago

How to Accelerate Infrastructure Provisioning for AI-Native Enterprises (2026 Guide)

Learn how AI-native enterprises can accelerate infrastructure provisioning with StackGen using Terraform modules, governance guardrails, AppStack templates, and AI-powered automation.

How to accelerate infrastructure provision for AI native enterprises (1)

Activity

@marrouchi added a new tool Hexabot , 1 week, 6 days ago.

Story Trending

@vaibhavgupta shared a post, 2 weeks ago

6+ Production Ready Shadcn Alert Components Collection

#shadcn ... #shadcn ... #shadcn ... #shadcn ... #shadcn ...

Looking for Shadcn Alert Components? Explore reusable alert patterns for SaaS apps, dashboards, onboarding flows, and admin interfaces.

Link

@radhika_rv16 shared a link, 2 weeks ago

SRE, Writer

Platform Engineering: From Bash Scripts to AI-Powered Platforms — and What Comes Next

I once got a Slack message at 11pm: "hey, prod seems slow?" No runbook. No alert. Just me, SSH'd into a box, squinting at top output. That was platform engineering. Here's how far we've come — and where AI is taking us next.

Course

@eon01 published a course, 2 weeks ago

Founder, FAUN.dev

GitOps the Hard Way, with Argo CD

#Helm #gitops #Argo CD #gitlab #kuberne...

Build Real GitOps Pipelines From Empty Clusters to Automated Deploys

Story Trending

@laura_garcia shared a post, 2 weeks, 1 day ago

Software Developer, RELIANOID

𝗥𝗘𝗟𝗜𝗔𝗡𝗢𝗜𝗗 𝗮𝘁 𝗟𝗼𝗻𝗱𝗼𝗻 𝗧𝗲𝗰𝗵 𝗪𝗲𝗲𝗸 𝟮𝟬𝟮𝟲

🚀 𝗝𝗼𝗶𝗻 𝗥𝗘𝗟𝗜𝗔𝗡𝗢𝗜𝗗 𝗮𝘁 𝗟𝗼𝗻𝗱𝗼𝗻 𝗧𝗲𝗰𝗵 𝗪𝗲𝗲𝗸 𝟮𝟬𝟮𝟲 📅 June 8–12, 2026 📍 London, United Kingdom London Tech Week brings together innovators, enterprises, startups, investors, and technology leaders to explore the future of AI, cybersecurity, cloud infrastructure, digital transformation, and emerging technologi..

Unsloth is an open-source toolkit for training and fine-tuning large language models faster and with less memory than a standard Hugging Face stack. Its core library replaces PyTorch's default autograd with custom backpropagation kernels written in OpenAI's Triton language, which is where most of its speed and memory savings come from. It supports LoRA, QLoRA, full fine-tuning, reinforcement learning, pretraining, and 4-bit, 16-bit, and FP8 training, across more than 500 text, vision, audio, and embedding models.

The practical draw is hardware reach. QLoRA workflows in Unsloth let you fine-tune an 8B model on a single 12 GB consumer GPU, and the project headlines roughly 2x faster training with about 70 percent less VRAM versus baseline implementations, though the exact figures vary by model, GPU, and config. A 2026 update added faster mixture-of-experts training, with models like Qwen3-30B-A3B fine-tunable on about 17.5 GB of VRAM. It runs on NVIDIA (including Blackwell and DGX Spark), AMD, and Intel GPUs, with free Colab and Kaggle notebooks for trying it without local hardware.

It fits cleanly into the local-AI workflow. Unsloth integrates with Hugging Face transformers and TRL, and uses llama.cpp to save and run models, exporting to GGUF for Ollama or LM Studio as well as safetensors. As of 2026 it also ships Unsloth Studio, a local no-code GUI that covers the full lifecycle from dataset creation to training to running and comparing GGUF and safetensors models, with tool-calling, web search, and an OpenAI-compatible API, all running offline on Mac and Windows, with the core library under the Apache 2.0 license.