Commit graph

15 commits

Author SHA1 Message Date
michaelxer
bdf4ec8b24
fix: fall back to /models probe when base URL returns 404 (#3205)
_ping_endpoint() probes the bare base URL for non-Ollama endpoints.
OpenAI-compatible servers like llama-swap return 404 on the /v1 prefix
but 200 on /v1/models, causing endpoints to appear offline despite being
fully functional.

Add a /models fallback when the base URL returns a non-auth 4xx.
Auth failures (401/403) are treated as definitive — probing /models
would just repeat the same rejection.

Fixes #3181

Co-authored-by: michaelxer <michaelxer@users.noreply.github.com>
2026-06-07 16:09:33 +02:00
Ocean Bennett
5911b8c0dc
fix(models): allow same endpoint URL with different keys (#2758)
* fix(models): allow same endpoint URL with different keys

* fix(models): show endpoint key fingerprints
2026-06-05 21:12:14 +02:00
Alexandre Teixeira
452a94fb1b
refactor(tests): centralize fake endpoint resolver cleanup
Test-only refactor continuing #2523. Centralizes the final repeated fake src.endpoint_resolver cleanup pattern into a focused import-state helper.
2026-06-05 13:23:46 +01:00
Alexandre Teixeira
3b292403dc
fix(tests): accept verify in endpoint HTTP mocks
Updates endpoint/model-route test HTTP mocks to accept the verify keyword argument passed by endpoint probing code. Restores one focused part of the Python CI baseline tracked in #2580.
2026-06-04 17:53:18 +01:00
Yuri
a2e691da2b
fix(models): stabilize proxy endpoint refresh behavior
* fix: support large proxy model endpoint refresh

Large OpenAI-compatible proxy endpoints can expose hundreds of models and make /v1/models slow. Treating those endpoints like local model servers caused model picker opens and background probes to repeatedly hit /models, producing timeouts and making otherwise usable endpoints appear offline.

Make model endpoint discovery cached-first for normal UI usage, add explicit proxy/API classification and refresh policy fields, exclude proxy/API endpoints from aggressive local probing, and preserve cached models when refresh fails.

Manual Test/Add/Refresh actions still fetch the full model list with longer timeouts so users can intentionally import large proxy model lists without blocking normal model picker usage.

* fix: preserve endpoint ping status semantics
2026-06-04 04:56:11 +01:00
pewdiepie-archdaemon
6861c41580 Reapply "Merge branch 'main' of github.com:pewdiepie-archdaemon/odysseus"
This reverts commit cc8fe2f6e3.
2026-06-03 22:47:00 +09:00
pewdiepie-archdaemon
cc8fe2f6e3 Revert "Merge branch 'main' of github.com:pewdiepie-archdaemon/odysseus"
This reverts commit 8161c1253d, reversing
changes made to 8c2705b42a.
2026-06-03 22:46:19 +09:00
Alexandre Teixeira
145f4fd2b4
feat(models): support pinned endpoint model IDs 2026-06-03 13:00:07 +01:00
red person
42ae905df7
fix(models): clear deleted endpoint fallback refs (#1207) 2026-06-02 23:41:04 +09:00
red person
76a7685105
fix(models): clear stale speech endpoint settings (#1196) 2026-06-02 23:32:01 +09:00
Hayk Arzumanyan
514050d098
Models: rewrite Docker loopback endpoints to host gateway
In Docker, a model-endpoint URL pointing at loopback (e.g. the LM Studio
default http://localhost:1234/v1) targets the Odysseus container itself, not
the host running the server, so the probe gets a connection error and the
endpoint is rejected with a misleading 'No models found for that provider/key'.
Rewrite loopback to host.docker.internal (which compose already maps to
host-gateway) for the probe and the saved URL, mirroring the existing Ollama
handling. Gated on actually being in a container with the gateway reachable, so
native installs and gateway-less deploys are untouched.

Fixes #25

Co-authored-by: Claude <noreply@anthropic.com>
2026-06-02 20:34:40 +09:00
Prakhya
a96593a99b
Improve Ollama endpoint error messages 2026-06-02 05:53:50 +09:00
Alexander Kenley
2c4b8b57dd
feat(ai): add OpenRouter and Ollama Cloud providers (#231)
Co-authored-by: Alex Kenley <Alex.Kenley@threatvectorsecurity.com>
2026-06-01 14:26:10 +09:00
Ranjan Sharma
058d32451c
Fix fresh checkout test failures
Make .env optional in tests and prevent endpoint resolver stubs from leaking into model route tests.
2026-06-01 02:22:17 +00:00
pewdiepie-archdaemon
e5c99a5eee Odysseus v1.0 2026-05-31 23:58:26 +09:00