ollama/llm
Daniel Hiltgen 909a88c5c0
Improve crash reporting (#7728)
Many model crashes are masked behind "An existing connection was forcibly closed by the remote host"
This captures that common error message and wires in any detected errors from the log.

This also adds the deepseek context shift error to the known errors we capture.
2024-11-19 16:26:57 -08:00
..
filetype.go Add support for IQ1_S, IQ3_S, IQ2_S, IQ4_XS. IQ4_NL (#4322) 2024-05-23 13:21:49 -07:00
ggla.go image processing for llama3.2 (#6963) 2024-10-18 16:12:35 -07:00
ggml.go refactor kv estimation 2024-11-01 16:23:55 -07:00
ggml_test.go llm: speed up gguf decoding by a lot (#5246) 2024-06-24 21:47:52 -07:00
gguf.go image processing for llama3.2 (#6963) 2024-10-18 16:12:35 -07:00
llm_darwin.go Optimize container images for startup (#6547) 2024-09-12 12:10:30 -07:00
llm_linux.go Optimize container images for startup (#6547) 2024-09-12 12:10:30 -07:00
llm_windows.go runner: Set windows above normal priority (#6905) 2024-09-21 16:54:49 -07:00
memory.go refactor kv estimation 2024-11-01 16:23:55 -07:00
memory_test.go Rename gpu package discover (#7143) 2024-10-16 17:45:00 -07:00
server.go Improve crash reporting (#7728) 2024-11-19 16:26:57 -08:00
status.go Improve crash reporting (#7728) 2024-11-19 16:26:57 -08:00