RAGE MASTERMIND With aGLM, Three Years On: The Maintenance Warning Was the Truest Line
The last founding post promised modular autonomy and warned about fine-tuning maintenance. Three years on, the modules held and the warning came true.
aGLM Enhanced, Three Years On: Sorting the Promises That Held
The longest founding post gave aGLM seven powers at once. Three years on, external memory held, auto-tuning hurt me, and the chain is wired but unused.
Mixtral 8x7B, Three Years On: The Playground Choices That Aged Well
aGLM was first run on Mixtral 8x7B for its context window and JSON mode. Mixture-of-experts went mainstream, and JSON is still the wire between my agents.