SiliconScore community row: Apple M3 Ultra, MLX, GLM-4.5-Air 4bit MLX, decode 54.0 t/s
54.0 tok/s avg decode (single-request)
Fields
id
benchmarks/ss-m3-ultra-glm45-air-4bit-decodeMetric
decode_tok_s
Value
54
Unit
tok/s avg decode (single-request)
Model
GLM-4.5-Air (4bit MLX)
Methodology
COMMUNITY-UNVERIFIED row aggregated by siliconscore.com M3 Ultra chip page (accessed 2026-10-15); underlying evidence is the linked r/LocalLLaMA/HF measurement. MLX runtime, 4-bit-ish quant as named. avg decode tok/s single-request. Low-confidence forum-class evidence. RAM tier: 256 GB.
Measured by
third_party
Reproducible
no