b10342 v0.1.13
ggml-org/llama.cppAug 10, 2026 → Aug 10, 20260 days between releases
Generating AI comparison
Analyzing what changed between b10342 and v0.1.13. This page updates automatically — usually within a minute.
The Two Releases
b10342
b10342
Aug 10, 2026· 2 months ago
Adds support for the new 'Granite-Switch' architecture, a dense all-attention model with LoRA adapters that can be selected per-token via control tokens using a router attention mechanism.
Full release notesv0.1.13
v0.1.13 — The grounding repair, a finance-math layer, and institutional research
Aug 10, 2026· 2 months ago
Fixes grounding issues preventing answers, introduces a finance-math layer (quantlib), adds a valuation engine, institutional research slash commands, and four new read-only data tools.
Full release notes