CLUSTER · TIER 3
llama.cpp adds external PLE loading and multi-shard model support
A llama.cpp pull request adds external PLE file loading, multi-shard model support, and conversion for PLE-enabled models. The changes support models including Gemma 4, Qwen 3.8 Next, and GLM 5.3 Flash.
Sources
1
X mentions
—
First seen
7Dago
Velocity
+2%/6h