Commit Graph

20 Commits

Author SHA1 Message Date
diegosouzapw
d4045e74b5 fix(models): honor per-model targetFormat for custom models (#2905)
Custom models (added via the UI) on opencode-go / openai-compatible nodes always
routed as OpenAI-compatible because there was no per-model targetFormat: addCustomModel
didn't accept it, the API schema stripped it, and getModelTargetFormat (static-registry
only) never saw it — so a custom model needing the Anthropic Messages shape fell back to
the provider default 'openai'.

Thread an optional targetFormat through addCustomModel / replaceCustomModels /
updateCustomModel + the providerModelMutationSchema + the POST/PUT route, surface it from
getModelInfo (one combined custom-model metadata lookup alongside apiFormat), and use it
in chatCore's targetFormat resolution before the provider default.

Closes #2905
2026-05-31 11:17:52 -03:00
墨林ObsidianGrove
83f242fa4d fix: clear synced provider models on delete all
Ensure provider-level delete all removes synced available model lists so imported models do not reappear after clearing managed models.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-15 01:39:31 +08:00
墨林ObsidianGrove
153a421354 test: cover managed model alias lifecycle
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-14 20:52:08 +08:00
墨林ObsidianGrove
bcbc095798 fix: sync managed model aliases with visibility
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-14 19:31:38 +08:00
Diego Rodrigues de Sa e Souza
37cc63e493 Release v3.6.0 (#1109)
* chore: create release/v3.6.0 branch

* fix: add row count limits to prevent DB bloat and handle HTML error responses (#1104)

Integrated into release/v3.6.0

* fix combo smoke test payload for thinking models (#1105)

Integrated into release/v3.6.0

* fix: improve Android/Termux ARM64 support for better-sqlite3 (#1107)

Integrated into release/v3.6.0

* chore: finalize CHANGELOG and sync versions for v3.6.0

* fix(tests): align CI tests with v3.6.0 changes

- compliance: match new cleanupExpiredLogs return shape (trimmed/maxRows)
- model-sync: accept masked email in account field
- e2e: allow 401/403/307 for auth-protected /api/providers endpoint

---------

Co-authored-by: diegosouzapw <diegosouzapw@users.noreply.github.com>
Co-authored-by: Paijo <14921983+oyi77@users.noreply.github.com>
Co-authored-by: Randi <55005611+rdself@users.noreply.github.com>
Co-authored-by: Suhayli <73960279+Suhay1i@users.noreply.github.com>
2026-04-10 09:56:42 -03:00
tombii
557509ef84 fix(model-sync): skip replace when auto-sync returns empty model list
Prevent auto-sync from wiping manually-imported models when the upstream
/models endpoint fails, times out, or returns an empty list. Added
`allowEmpty` option (default false) to replaceCustomModels — callers that
intentionally clear all models (DELETE ?all=true) pass `allowEmpty: true`.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-01 23:01:30 +02:00
R.D.
8333f3d9de feat: add "Clear All Models" button on provider detail page
Adds a button next to the Auto-Sync toggle to clear all custom models
for a provider. Extends DELETE /api/provider-models to support ?all=true
parameter for bulk deletion.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-25 21:11:17 -04:00
zhang-qiang
a8ca88797a feat(dashboard,sse,api): per-model upstream headers, compat PATCH, chat alignment
- Store/sanitize upstreamHeaders; shared forbidden header names (upstreamHeaders.ts)

- chatCore: buildUpstreamHeadersForExecute; T5 recomputes; 401 retry uses translatedBody.model

- Dashboard compat popover + i18n; Zod partialRecord + header value newline guard

- Executors merge upstreamExtraHeaders; sanitize unit tests

- Dev: bootstrap env in run-next, instrumentation-node import, credentialLoader dedupe

Made-with: Cursor
2026-03-24 17:24:11 +08:00
zhang-qiang
a2d7cbe8fe feat(compat): per-protocol model compatibility config (V5)
Add per-protocol compatibility options (compatByProtocol) allowing users to configure normalizeToolCallId and preserveOpenAIDeveloperRole per client request protocol (OpenAI Chat, Responses API, Anthropic Messages) instead of globally. Includes frontend Map lookup optimization, type safety improvements, and client-safe constant extraction.

Made-with: Cursor
2026-03-21 15:23:42 +08:00
zhang-qiang
1a099ea2f2 feat(zws-v2): model compat, provider-models hardening, provider page types
- roleNormalizer/translator: ZWS v2 role handling and comments

- models + schemas: compat overrides, nullable preserveOpenAIDeveloperRole

- provider-models API: generic GET 500; compatOnly validates known provider

- providers [id] page: typed props; minimal saveModelCompatFlags PATCH

Made-with: Cursor
2026-03-20 23:03:52 +08:00
zhang-qiang
13c45807ef feat: protocol-scoped model compat (V3)
- compatByProtocol per openai/openai-responses/claude

- getters take sourceFormat; chatCore passes it

- UI: protocol selector in compat popover, dark mode select

- shared/constants/modelCompat for client-safe import (fix node:crypto build)

- ZWS_README_V3.md

Made-with: Cursor
2026-03-20 22:06:03 +08:00
zhang-qiang
a7fe369ea0 fix: resolve role param error for Responses API + MiniMax (developer→system)
- Add preserveDeveloperRole option and model compat override

- Normalize developer→system in roleNormalizer when not preserving

- Translator runs normalizeRoles for Responses API with option

- UI: ModelCompatPopover with do not preserve developer toggle

- Add ZWS_README_V2 documenting cause and fix

Made-with: Cursor
2026-03-20 21:06:10 +08:00
diegosouzapw
601cc21a44 feat: call log response content, per-model tool call ID, key PATCH & validation (#470) 2026-03-19 08:41:01 -03:00
jack
6cad4fae8e fix(providers): persist supported endpoints with explicit save for custom models 2026-03-11 11:20:25 +00:00
diegosouzapw
228ebf436e feat: endpoint-aware model management + fix 3 bugs (#212, #213, #200)
Bug Fixes:
- #212: Auto-generate API_KEY_SECRET at startup (like JWT_SECRET)
- #213: Circuit breaker now scoped per-model instead of per-provider
- #200: Connectivity fallback for custom providers (Ollama, LM Studio)

Features:
- #204: API Format selector (Chat Completions / Responses API) for custom models
- #205: Combo endpoint field (chat / embeddings / images) in schema
- #206: Supported Endpoints checkboxes (chat, embeddings, images, audio)
- Custom models with endpoint tags appear in /v1/embeddings and /v1/images/generations
- Model catalog includes api_format, type, and supported_endpoints metadata
- Provider detail page shows badges for non-default endpoint configurations

Files changed: instrumentation.ts, combo.ts, validation.ts, models.ts,
schemas.ts, provider-models/route.ts, providers/[id]/page.tsx,
catalog.ts, embeddings/route.ts, images/generations/route.ts
2026-03-05 18:49:07 -03:00
diegosouzapw
751ff77b7c fix: extract validation helpers to fix webpack barrel-file resolution bug
Extracted validateBody, isValidationFailure, and loginSchema from the
935-line schemas.ts barrel file into a dedicated helpers.ts module.
Updated 70 API route files to import directly from helpers.ts.

Root cause: webpack on certain environments fails to resolve exports
from the bottom of large barrel files (schemas.ts), causing
'(0, O.Jb) is not a function' errors in production builds.

Fix: Split validation helpers into a small dedicated module (helpers.ts)
so webpack can correctly resolve all exports regardless of file size.

- TypeScript compiles with 0 errors
- All API routes updated to import from helpers.ts
- schemas.ts re-exports from helpers.ts for backward compatibility
2026-03-05 08:05:58 -03:00
diegosouzapw
3510d8c0bc feat: add A2A protocol support and refactor API validation layer
Implement Agent-to-Agent (A2A) JSON-RPC 2.0 endpoint with smart routing
and quota management skills, SSE streaming, and task lifecycle management.

- Add /a2a route with message/send, message/stream, tasks/get, tasks/cancel
- Add /.well-known/agent.json agent card endpoint
- Introduce Zod-based request validation schemas for all v1 API routes
- Extract shared getUnifiedModelsResponse to reduce duplication across
  /models, /v1/models, and /a2a model listing
- Refactor chat, embeddings, moderations, and models routes to use
  centralized validation and error handling
- Add A2A task manager, routing logger, and streaming utilities
2026-03-04 18:52:25 -03:00
nyatoru
4ea0426034 feat: extract shared auth utility a cooldown/availability checksnd fix custom provider model resolution 2026-02-24 21:19:57 +07:00
nyatoru
243cc4b60b feat: add authentication to alias API and improve model save error handling 2026-02-24 14:28:51 +07:00
diegosouzapw
71d14209a4 feat: OmniRoute v1.0.0 — Intelligent AI Gateway & Universal LLM Proxy
OmniRoute is an intelligent API gateway that unifies 20+ AI providers behind a single
OpenAI-compatible endpoint. Features include intelligent routing with 6 strategies,
multi-format translation (OpenAI/Claude/Gemini/Responses API), circuit breakers,
semantic caching, combo fallback chains, real-time health monitoring, and a full
dashboard with provider management, analytics, and CLI tool integration.

Key highlights:
- 20+ providers (Claude Code, Codex, Gemini CLI, GitHub Copilot, iFlow, Qwen, Kiro, etc.)
- 6 routing strategies (Fill First, Round Robin, P2C, Random, Least Used, Cost Optimized)
- Export/Import database backup with full archive support
- Translator Playground with 4 modes (Playground, Chat Tester, Test Bench, Live Monitor)
- 100% TypeScript across src/ and open-sse/
- Docker support with multi-stage builds
- Comprehensive documentation and 9 dashboard screenshots
2026-02-18 00:02:15 -03:00