M3 numbers look good but am gonna wait for the weights before I comment any further
@theahmadosman
-
Open weight models lag behind closed models but shrinking
By
–
Open weight models have lagged the state of the art closed models by four months The lag is shrinking not expanding unlike what paid influencers would like you to believe
-
Pro Tip for Agents: Modularity, DDD, Explicit Specs, Documentation
By
–
PRO TIP For Codex Cli & other agents like Claude Code, Droid, OpenCode, etc There’s a crucial recipe: 1. Modularity 2. Domain-Driven Design 3. Painfully explicit specs 4. Excessive documentation This is systems engineering I’m not saying agents aren’t
-
Opus 4.8: ‘You’re absolutely right’ in the worst way
By
–
Opus 4.8 is "You’re absolutely right” in the worst possible ways
-
124GB NVFP4 checkpoint with FP8 KV Cache
By
–
They provided an NVFP4 checkpoint, 124GB and runs w/ FP8 KV Cache
-
GPT 5.5 beats local models; Opus 4.6 superior to 4.7/4.8
By
–
– GPT 5.5 beats it – Nothing local comes close to GPT 5.5 yet – Opus 4.6 is better than Opus 4.7 / 4.8 I have said all of that before, dunno why you're being so aggressive
-
Testing Opus 4.8 Model Performance in Different Harnesses
By
–
Tried Opus 4.8 in different harnesses than Claude Code and nah this model is just stupid, good job Anthropic lol
-
Prioritizing local AI engagement over selling courses or grifting
By
–
I would rather get people invested and interested in Local AI than sell them this content or grift a course Not even pointing the urls to a website but X threads to avoid anyone telling me I am making money by showing ads or whatever lol
-
Local AI Setup Guide: Hardware, Software, Model Mechanics
By
–
DROP EVERYTHING Everything you need to get started with Local AI completely FOR FREE Hardware. Software. Anything in between. > TheLocalAIBook DOT com Local LLMs From Zero to Hero Articles – Hardware foundations
– Software stacks
– Model mechanics > BuyAGPU dot AI The Buy -
8B MoE Model Trained for Agentic Local Workflows
By
–
8B MoE (1B Activated) trained on 38 trillion tokens for local and agentic workflows