ollama/ml/nn
Michael Yang de82b1f9a3 cleanup attention interface
the updated interface supports variadic attention options which
removes the need for individual `AttentionWith...` functions. it means
more models can use the attention interface, e.g. models with
custom masks, logit softcapping, etc.

additionally, this interface should be less error prone since there are
now reasonable defaults for all optional parameters
2025-12-16 09:45:04 -08:00
..
attention cleanup attention interface 2025-12-16 09:45:04 -08:00
pooling chore: update models to use slice/chunk/chunksections (#12934) 2025-11-13 15:20:12 -08:00
rope fix: qwen2.5 vl rope (#13486) 2025-12-15 17:30:33 -08:00
attention.go cleanup attention interface 2025-12-16 09:45:04 -08:00
convolution.go fix: conv2d bias (#12834) 2025-10-29 11:03:43 -07:00
embedding.go next ollama runner (#7913) 2025-02-13 16:31:21 -08:00
linear.go update vendored llama.cpp and ggml (#11823) 2025-08-14 14:42:58 -07:00
normalization.go next ollama runner (#7913) 2025-02-13 16:31:21 -08:00
rope.go refactor rope 2025-12-08 14:42:22 -08:00