I've been testing IBM's Granite4.2-3B model on my M2 MacBook Air, and comparing with…
This is a AI post classified by Jev as Open & local models (a model release), kept by the AI Radar because it carries real work, not commentary.
I've been testing IBM's Granite4.2-3B model on my M2 MacBook Air, and comparing with other viable small models. Looking to see what makes the best overall 'brain' for an on-device personal agent experience. All models tested with 4-bit, and 32k context window. - MiniCPM5-2B is still the speed king. That's the snappy back and forth that really surprises you on such old hardware. - For nuanced/vague context retrieval, this is where Granite falls behind. The others are all relatively comparable with each other - within tolerance. - With tool use protocols and related accuracy, all performed n
Posted by GooGZ AI (1.4k followers) 2 days ago · 13 likes · 402 views · view the original post on X. Kept by the AI Radar as Open & local models.
More AI work like this
- dear rtx 3060 owners, and every 12gb card behind it. bonsai 2 27b dense went from 26 to… — @sudoingX
- Nimble is able to process images! — @madiator
- Woah...it claims to be open Jev. And it's 100% opensource. — @Saboo_Shubham_
- your rtx 3060 was running bonsai 2 at 26 tok/s this morning and now does 40 tok/s, i… — @sudoingX
- Meet DeepSeek-V4.1-Flash: a blazing fast image-text-to-text model. It takes both images… — @HuggingModels
- Meet Edge0-35B-A3B-preview: a 35B parameter MoE model built for edge inference. It runs… — @HuggingModels
- Putting Jeff-1 out into the world, this is an attempt at a Jev like model, built on qwen… — @DJLougen
- Train a 9M parameter language model from scratch in five minutes. — @tom_doerr
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. 27.2k posts from 4.8k X accounts over the last 14 days, 1.2k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-19 18:50 UTC. Full method.