GLM 5.2 on Mac Studio Speedup PR
Summary
GLM 5.2 delivers major performance gains on Mac Studio with 512GB RAM, achieving prefill speeds above 100 t/s at high context lengths and enabling 4-bit quantization for contexts over 100k tokens, as detailed in a pull request by the oMLX creator.
Similar Articles
@AlexFinn: I can't believe this is real I have GLM 5.2 running 100% locally on my Mac Studio. 2 bit quant. The results I'm getting…
A user reports running GLM 5.2 locally on a Mac Studio with 2-bit quantization, claiming it outperforms Opus 4.8 and enables free, private superintelligence for coding and agent tasks.
GLM 5.2 on consumer hardware
A user tested the unsloth quantized GLM-5.2 model on a high-end consumer-like system with dual RTX 5090, achieving 12 tokens per second.
GLM 5.3 Flash Q4 @ 60tps / 550tps on M3 Ultra
Optimizations for GLM 5.3 Flash on Apple M3 Ultra achieve up to 550 t/s prefill and 38 t/s inference speed through kernel fusion and efficient memory use, without quality loss.
Running GLM 5.2 on 4xGB10 with a 100G Switch, 330k ctx, ~25 t/s tg, ~650 t/s pp
This post details running GLM 5.2 on a 4xGB10 setup with a 100G switch, achieving ~25 tok/s decode and ~650 tok/s prefill at 330k context. It includes hardware costs, performance benchmarks with Depth Prefill, and notes on model pruning for longer context.
@cryptoresetlife: 本地无审核版 GLM5.2 754B 参数模型 231GB 在我的MAC studio M3 ultra 512gb 上部署成功了 @support_huihui 948 tokens / 4分25秒 = 948 / 265 ≈ 3.6 …
An uncensored version of the GLM5.2 754B parameter model (231GB GGUF) was successfully deployed on a Mac Studio M3 Ultra with 512GB RAM, achieving approximately 3.6 tokens/s.