llamaperf

Tiel-Coder 35B (3B active) AX , native decode / python-lru

on M3 Ultra 512GB · AX Engine (574d2d19 + 5e07befe; MLX 0.32.2)

Apple hardware
Oct 11, 2026
Throughput
201.3 t/s gen
Quant
AXQ-MXFP4 (MLX)
MTP (Multi-Token Prediction)
on
System RAM
512 GB

Use cases

coding

Summary

User reports Tiel-Coder 35B-A3B at 201.3 t/s median native decode on an Apple Mac Studio M3 Ultra with 512 GiB unified memory. Setup is AX Engine with the AXQ-MXFP4 MLX export, explicit throughput-MTP at maximum draft depth three, expert streaming off, and cold request KV with no prefix reuse. The figure is native decode only on a python-lru prompt with 194 input tokens and 256 total output tokens; completion throughput including TTFT is 160.35 t/s and median native TTFT is 329.74 ms.