llama.cpp: changelog & breaking changes

Auto-tracked from github.com/ggml-org/llama.cpp. Updated Jul 06, 2026.

b9893

2026-07-07
âš  Breaking changes
  • opencl: remove stray token in flash_attn_f32_f16.cl

b9892

2026-07-06

✓ No breaking changes detected — looks safe to upgrade.

b9891

2026-07-06

✓ No breaking changes detected — looks safe to upgrade.

b9890

2026-07-06
âš  Breaking changes
  • CUDA: remove -sm row, refactor cuBLAS ()
  • CUDA: remove -sm row, refactor cuBLAS

b9888

2026-07-06

✓ No breaking changes detected — looks safe to upgrade.

b9886

2026-07-06

✓ No breaking changes detected — looks safe to upgrade.

b9885

2026-07-06

✓ No breaking changes detected — looks safe to upgrade.

b9884

2026-07-06

✓ No breaking changes detected — looks safe to upgrade.

Get breaking-change alerts for YOUR stack

Drop your email + the libraries you use. We'll alert you when one ships a breaking change — before it breaks your deploy. Free.

⚡ Want an alert the moment your stack breaks — not a weekly digest? ChangeRadar Pro adds instant alerts + unlimited libraries + AI migration steps.Go Pro — $9/mo

Related breaking-change trackers

fastapichangelog & breaking changes next.jschangelog & breaking changes numpychangelog & breaking changes pytorchchangelog & breaking changes reactchangelog & breaking changes tailwindcsschangelog & breaking changes TypeScriptchangelog & breaking changes lernachangelog & breaking changes