GLM-5.2-FP8

As models, contexts, and workloads grow, hidden assumptions in inference infrastructure can surface as output anomalies. Reliability requires more than throughput, latency, and availability. It also r

As models, contexts, and workloads grow, hidden assumptions in inference infrastructure can surface as output anomalies. Reliability requires more than throughput, latency, and availability. It also requires preserving the correctness of model state behind every generation.

After fixing correctness issues, we turned to the next bottleneck: Prefill throughput and GPU memory pressure in long-context Coding Agent serving. To address this, we introduced LayerSplit, a layer-w

After fixing correctness issues, we turned to the next bottleneck: Prefill throughput and GPU memory pressure in long-context Coding Agent serving. To address this, we introduced LayerSplit, a layer-wise KV Cache storage scheme. Instead of duplicating all layers on every GPU, https://t.co/OGptVovbtf

Fantastic to see GLM being applied to such fresh, dynamic scenarios.