GLM 5.2 Local Benchmark Tested
gutard · reddit · 2026-07-14
A user ran GLM 5.2 on a MacBook Pro M5 48GB using Flash MoE, sharing local inference benchmark results.
- With pre-fill (approx. 4.4k tokens), it took 6 minutes and 38 seconds to generate 501 tokens, at a speed of about 2t/s.
- Without pre-fill (400 input tokens, 337 output tokens), it took 3 minutes and 17 seconds.
- A speed of 2.8t/s was also observed, though it fluctuates with the machine's current load.
They mentioned using pipenetwork/GLM-5.2-MLX-mixed-36bit, which is about 332GB in size. Next, they plan to test having it plan tasks overnight and execute them using Qwen during the day.
More from Infra
- SpaceX strikes $40B deal with Apollo Global to lease AI chips and data center hardware — Polymarket · 2026-10-26
- India Needs at Least 5GW of Compute by 2028 to Stay Relevant, User Argues — EAccelerate_42 · 2026-10-26
- SpaceX AI revenue up 830%, Claude designs 318K antivenom proteins in 64 hours — PeterDiamandis · 2026-10-26
- Meta plans up to $95 billion stock sale to fund AI buildout — The Decoder · 2026-10-26
- Pandora M1: first prototype of a physical device for distributed GPU monitoring — tarantulae · 2026-10-26
- INT21's AI agent swarm rewrote Drama 3's inference runtime in Rust within one hour, beating SGLang-Omni — bingxu_ · 2026-10-26