All posts
-
Qwen 3.8 Flash Next and the on-disk n-gram table
The on-disk n-gram table is truly a magic wand for performance
-
Reflecting on 4 years of LLM software development evolution
Shifting left and how LLM coding feels like the final evolution of DevOps
-
More FP4-local testing with GLM-5.3
Trying Nvidia's next local optimized model
-
Testing Nvidia's new FP4-local quant of Qwen 3.8 Flash Next
An incredible model specifically designed for single DGX Spark or RTX 6000 Pro setups.
-
Benchmarking pruned GLM 5.3
Testing autotrusts new pruned GLM 5.3 to see if it can become my local darling
-
Personal apps vol 41237
Deploying a personal search frontend for privacy focused searching
-
Qwen 3.8 Flash Next vs MiMo 2.6 Flash
Sadly for a 2x DGX Spark setup MiMo isnt the next great model
-
Writing software is our generations blacksmithing
Will people value handmade code the way we do handmade knives?
-
Another weekend spent building personal apps
Introducing RSSgate - AI powered news reader
-
Keeping up with open-source AI
Hopefully a helpful resource for others trying to follow new model releases
-
Monitoring an LLM fixing the monitoring for monitoring the LLM
Just a little bit of fun.
-
Cinematic start to Fall
My favorite season is here again.
-
LLM tuning LLM
Having Claude Opus 5.5 tune vLLM and Qwen 3.8 for my actual usage
-
Sekka - a TUI chatbot client
A mornings fun with personal app building and energy tracking
-
The rising cost of AI sovereignty
With datacenters eating up all the fabrication capacity AI sovereignty is becoming out of reach
-
ChatGPT integrated into Siri is embarrassingly bad
Remember Apple Intelligence at WWDC 2024?
-
FP8 compute and you
Primer on FP8 and the iPhone 18 Pro / A20 Pro
-
Cognition vs Determinism
Musing on agent design patterns for success, safety, and cost
-
Qwen 3.8 Flash Next is scary good
A model that runs well on 120GB of RAM that covers 90% of my needs
-
Sub-frontier LLMs are super exciting again
A long weekend playing with the smaller "Flash" releases from Deepseek, Qwen, and GLM
Tags: cto6 games2 other1 photography8 technology27 tinkering16