root d935339280 fix: report actual profile context size in /api/show probe endpoint ... Hermes Desktop reads the context size from /api/show's 'parameters' field. This was hardcoded to 'num_ctx 4096' for every model, causing 'context too small' errors when the user's system prompt + conversation exceeded 4K tokens. Now extracts the actual ctx-size from the profile's flags and returns the correct value (e.g. 'num_ctx 131072' for the 128K profiles).		2026-06-16 21:04:40 +00:00
.hermes/plans	fix: add probe endpoints and no-model fallback for Hermes Desktop compatibility	2026-06-15 15:22:15 +00:00
deploy	fix: convert underscores to hyphens in llama-server flag names, fix n_ctx→ctx-size rename	2026-06-16 20:54:32 +00:00
docs	Added next changes	2026-06-15 00:09:31 +00:00
scripts	feat: add sync_models.py script to auto-update Hermes custom_providers from router model list	2026-06-15 21:10:36 +00:00
sidecar	fix: convert underscores to hyphens in llama-server flag names, fix n_ctx→ctx-size rename	2026-06-16 20:54:32 +00:00
tests	fix: change sidecar port from 8081 to 8080	2026-06-15 13:17:31 +00:00
.env	.env	2026-06-09 13:57:22 +03:00
.gitignore	Epic: Model Switching via Sidecar — Issues #2-#3	2026-06-15 00:49:24 +00:00
CONTEXT.md	Epic: Model Switching via Sidecar — Issues #4-#7 + #8 deployment	2026-06-15 01:13:36 +00:00
docker-compose.yml	fix: resolve port conflict between sidecar and llama-server	2026-06-15 15:31:31 +00:00
Dockerfile	Initial commit: migrate intelligence-router files	2026-06-09 11:48:43 +01:00
main.py	fix: report actual profile context size in /api/show probe endpoint	2026-06-16 21:04:40 +00:00
pytest.ini	feat: add 15 model profiles to manifest.yaml	2026-06-15 12:34:46 +00:00
requirements.txt	Epic: Model Switching via Sidecar — Issues #4-#7 + #8 deployment	2026-06-15 01:13:36 +00:00