I am finally learning how to fine-tune models using @UnslothAI doing a "hello world" test.
This is a AI post classified by Jev as AI infra & evals (a tutorial), kept by the AI Radar because it carries real work, not commentary.
I am finally learning how to fine-tune models using @UnslothAI doing a "hello world" test. Base model -> unsloth/Qwen3-4B-Instruct-2507 It's a 4B params, ~8GB on disk. It's a good model for learn as you can train it relatively fast and see results immediately. training data -> mlabonne/FineTome-100k It's 100k ChatML conversations, cleaned and curated (a well-known community SFT set). That's what I am "teaching" the model: high-quality Q&A / explanation-style conversations. What's LoRA? That's the core trick. Instead of updating all 4B weights (needs ~4×8GB+ for gradients/optimizer - won
Posted by Javier • priv/acc (4.4k followers) 1 h ago · 5 likes · 144 views · view the original post on X. Kept by the AI Radar as AI infra & evals.
More AI work like this
- Thunderobot launches Ryzen AI Max+ 395 laptop with 85GB SSD AI cache — @VideoCardz
- Nearly three years ago, Rendair started with just one serverless endpoint on Runpod. — @runpod
- Stay on top of model changes, low credits, workspace budgets, and API key limits with… — @OpenRouter
- BeeLlama v0.4.7 Preview just made DFlash 2 by @inco_ai much more efficient in terms of… — @Anbeeld
- Me looking at "Apple's M5 Ultra Mac Studio Shines in Local AI Tests" trending on X with… — @ivanfioravanti
- Your users have already generated a better eval dataset than you'd write yourself. — @sentry
- Qwen3.8-2.4T on vLLM: a Pareto frontier spanning 5K total tokens/s/GPU at high… — @vllm_project
- 🚀 Introducing Luminary, our AI wind tunnel for enterprise model selection. — @MultiverseCompu
Every post is read and classified by Jev (TypeSafe): what it is, which market it belongs to, and whether the link is a real tool. 39.3k posts from 5k X accounts over the last 14 days, 1.6k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-21 17:06 UTC. Full method.