ISTA coding prune of Flash-Next. Street GSQ-RCO quote on M5 Max. Bit width not in the quote.
Params
Flash-Next, 256/512 experts
Fastest decode
51 tok/s
M5 Max 64GB · ISTA GSQ-RCO · llama.cpp
Refusal
Official
Setups
1
Best reported pair · not live serving speed ↗
Runs on
Quoted one-stream decode on real boxes.
No public weights link on this desk yet.