llamaperf

KAT-Coder V2.5-Dev

on Unknown GPU · llama.cpp

Tone: positive
Oct 16, 2026
Throughput
14.0 t/s gen
Quant
Q4_K_M (GGUF)

Use cases

coding

Summary

User reports KAT-Coder V2.5-Dev at 14 t/s on an 8-core CPU with no GPU. Setup is llama.cpp with Q4_K_M GGUF weights. The user finds the result impressive for a small model on a potato iGPU and pure CPU, completing a solar system demo in under 10 minutes.