b10338 b10336
ggml-org/llama.cppAug 10, 2026 → Aug 10, 20260 days between releases
Generating AI comparison
Analyzing what changed between b10338 and b10336. This page updates automatically — usually within a minute.
The Two Releases
b10338
b10338
Aug 10, 2026· 2 months ago
Critical bug fix for the model saver that was incorrectly saving MoE (Mixture of Experts) model parameters, causing models like Qwen2MoE to fail loading after a save/load cycle.
Full release notesb10336
b10336
Aug 10, 2026· 2 months ago
Refactored WebGPU (WGSL) files to simplify the Flash Attention implementation, improving code maintainability and compilation efficiency.
Full release notes