fix: propagate throws from LanguageModel.newCache and adapt to mlx-swift-lm PR #62 #452

Job	Run time
build_and_unit_test	20m 14s
dflash-speculative-decoding	11m 3s
ssd-draft-memory-guard	1m 25s
speculative-decoding	12m 27s
speculative-decoding-eval	15m 49s
integration_matrix (graph)	45s
integration_matrix (vision, mlx-community/Qwen2-VL-2B-Instruct-4bit LiquidAI/LFM2.5-VL-450M-MLX-4...	2m 12s
integration_matrix (fixtures)	1m 31s
integration_matrix (audio, mlx-community/gemma-4-e4b-it-4bit)	3m 15s
integration_matrix (server, mlx-community/Qwen2.5-0.5B-Instruct-4bit)	58s
integration_matrix (server, -gemma-e2b, mlx-community/gemma-4-e2b-it-4bit, mlx-community/gemma-4-...	4m 36s
integration_matrix (contract, mlx-community/Qwen2.5-0.5B-Instruct-4bit)	1m 57s
integration_matrix (opencode, mlx-community/gemma-4-e4b-it-4bit)	4m 27s
integration_matrix (omni, mlx-community/gemma-4-e4b-it-4bit)	4m 14s
	1h 24m 53s

Navigation Menu