OpenSource/ollama @ pdevine/convert-cohere2

Jeffrey Morgan 1deafd8254 llama: update vendored code to commit 46e3556 (#8308)		3 mēneši atpakaļ
..
filetype.go	fda1e6b563 llm: bring fileTypes into alignment with llama.cpp (#7819)	5 mēneši atpakaļ
ggla.go	c7cb0f0602 image processing for llama3.2 (#6963)	6 mēneši atpakaļ
ggml.go	1bdab9fdb1 llm: introduce k/v context quantization (vRAM improvements) (#6279)	5 mēneši atpakaļ
ggml_test.go	cb42e607c5 llm: speed up gguf decoding by a lot (#5246)	10 mēneši atpakaļ
gguf.go	c7cb0f0602 image processing for llama3.2 (#6963)	6 mēneši atpakaļ
llm_darwin.go	cd5c8f6471 Optimize container images for startup (#6547)	7 mēneši atpakaļ
llm_linux.go	cd5c8f6471 Optimize container images for startup (#6547)	7 mēneši atpakaļ
llm_windows.go	dbba73469d runner: Set windows above normal priority (#6905)	7 mēneši atpakaļ
memory.go	63269668c0 Prevent underflow when FreeMemory < overhead (#8014)	4 mēneši atpakaļ
memory_test.go	abfdc4710f all: fix typos in documentation, code, and comments (#7021)	4 mēneši atpakaļ
server.go	1deafd8254 llama: update vendored code to commit 46e3556 (#8308)	3 mēneši atpakaļ
server_test.go	2ddc32d5c5 llm: do not error on "null" format (#8139)	4 mēneši atpakaļ
status.go	909a88c5c0 Improve crash reporting (#7728)	5 mēneši atpakaļ