v0.33.3-rc0: llama.cpp: version bump b10729 (#18160)
- Ollama: 22 events in the last 90 days
- Ollama: 20th Release in the last 90 days
- Previous: 4 days earlier · v0.33.2
What happened
llama.cpp: version bump b10729 Regenerate the compat hooks patch for b10729: upstream removed the whole-tensor load_data_for read (last consumer was llama-quantize, which now reads slabs via load_data_range). Keep the existing hook surface (constructor, skip loops, load_all_data, mtmd/clip) unchanged and add maybe_load_text_tensor_range, which materializes a text load op's output once per tensor and serves the new (…
Summary assembled by rule from the sources below
Why it's spreading
Sources
Release
- v0.33.3-rc0: llama.cpp: version bump b10729 (#18160)Ollama Releases