OpenSource/ollama

作者	SHA1 備註	提交日期
Jesse Gross	0c220935bd input: Rename Options to Batch	1 月之前
Jeffrey Morgan	da0e345200 ml: use input context for extracting outputs (#9875)	1 月之前
Jesse Gross	282bfaaa95 ollamarunner: Use a separate context per multimodal input	1 月之前
Jesse Gross	9679f40146 ml: Allow models to constrain inputs to a single batch	1 月之前
Michael Yang	5e2e0b46b1 fix: error if image requested without vision model	1 月之前
Michael Yang	63a394068c use 2d pooling	1 月之前
jmorganca	11bfa62796 add trailing \n\n after <end_of_image> to match reference implementation	1 月之前
jmorganca	f63e62e546 reduce kernel size, add TODO for loading from config	1 月之前
jmorganca	65b0f329d1 Revert "Allow models to force a new batch"	1 月之前
Jesse Gross	06007c0a18 Allow models to force a new batch	1 月之前
Jesse Gross	2c40c4d35e Fix follow up images and images split across batches	1 月之前
Michael Yang	6b32a2d549 compat with upstream gguf	1 月之前
Michael Yang	46bb0169c4 update model	1 月之前
Michael Yang	8934324b72 use fast attention	1 月之前
Michael Yang	0df1800436 set non-causal attention	1 月之前
Michael Yang	4b037a97dc add gemma vision encoder	1 月之前
Patrick Devine	5f74d1fd47 gemma2 impl	2 月之前