b10536 b10534
ggml-org/llama.cppAug 21, 2026 → Aug 21, 20260 days between releases
Generating AI comparison
Analyzing what changed between b10536 and b10534. This page updates automatically — usually within a minute.
The Two Releases
b10536
b10536
Aug 21, 2026· 2 months ago
Adds a lazy-loading feature to the server router that delays the loading of startup models until after the main server setup is complete.
Full release notesb10534
b10534
Aug 21, 2026· 2 months ago
Optimizes CUDA decoding performance by tuning the crossover point between mul_mat_vec_q and MMQ batch kernels, adding hardware-specific switch points to handle different GPUs and quantization types.
Full release notes