omni-inference
OpenAI 互換の中核推論エンドポイント(チャット補完・埋め込み・画像生成・音声の TTS/STT・モデレーション・リランク・Responses API)の仕様を、認証方法やリクエスト/レスポンス形式、プロバイダ別ルーティングとあわせて整理したスキルです。単一の安定したエンドポイントから複数のモデル提供元を呼び出せるため、開発者や AI エージェントはベンダーごとの独自アダプタを書く手間を省き、実装を移植しやすく保てます。
npx skills add https://github.com/cheaper-inference/omniroute --skill omni-inferenceBefore / After 効果比較
2 组チャット・埋め込み・画像・音声を同時に提供しようとすると、提供元ごとにドキュメントを読み込みリクエスト形式を自作する必要があり、2〜3社をつなぐだけで丸2日近くかかっていました。
この Skill は OpenAI 互換のエンドポイントとパラメータ規約で全推論機能を説明するため、同じリクエスト構造のまま複数ベンダーへ接続でき、作業は半日以内に収まります。
Overview
The core OpenAI-compatible inference endpoints: chat completions, embeddings, images, audio (TTS/STT), moderations, rerank, and the Responses API. The primary integration surface for AI agents.
Authentication
All requests require a valid Bearer token or session cookie. Obtain a token via POST /api/auth/login or configure REQUIRE_API_KEY=false for local development.
Endpoints
POST /api/v1/session-leasesGET /api/v1/searchPOST /api/v1/searchPOST /api/v1/chat/completionsGET /api/v1/wsPOST /api/v1/providers/{provider}/chat/completionsPOST /api/v1/api/chatPOST /api/v1/messagesPOST /api/v1/messages/count_tokensPOST /api/v1/responsesPOST /api/v1/embeddingsGET /api/v1/multimodal-embeddingsPOST /api/v1/multimodal-embeddingsPOST /api/v1/providers/{provider}/embeddingsPOST /api/v1/images/generationsPOST /api/v1/providers/{provider}/images/generationsPOST /api/v1/audio/speechPOST /api/v1/audio/transcriptionsPOST /api/v1/moderationsPOST /api/v1/rerankGET /api/v1GET /api/v1/providers/{provider}/modelsGET /api/v1/management/proxy-subscriptionsPOST /api/v1/management/proxy-subscriptionsGET /api/v1/management/proxy-subscriptions/{id}PATCH /api/v1/management/proxy-subscriptions/{id}DELETE /api/v1/management/proxy-subscriptions/{id}GET /api/v1/management/proxy-subscriptions/{id}/nodesPOST /api/v1/management/proxy-subscriptions/{id}/refreshPOST /api/v1/ocrPOST /api/v1/audio/translationsGET /api/v1/voicesPOST /api/v1/speech-to-textPOST /api/v1/text-to-speech/{voiceId}GET /api/v1/explain/routingGET /api/v1/providers/suggested-modelsGET /api/v1/provider-plugin-manifestGET /api/v1/{omnirouteCatchAll}POST /api/v1/{omnirouteCatchAll}PUT /api/v1/{omnirouteCatchAll}PATCH /api/v1/{omnirouteCatchAll}DELETE /api/v1/{omnirouteCatchAll}GET /api/v1/accounts/{id}/limitsPUT /api/v1/accounts/{id}/limitsGET /api/v1/agents/credentialsPOST /api/v1/agents/credentialsGET /api/v1/agents/healthGET /api/v1/agents/tasksPOST /api/v1/agents/tasksDELETE /api/v1/agents/tasksGET /api/v1/agents/tasks/{id}POST /api/v1/agents/tasks/{id}DELETE /api/v1/agents/tasks/{id}POST /api/v1/antigravityGET /api/v1/auto-combo/{channel}/candidatesGET /api/v1/batchesPOST /api/v1/batchesGET /api/v1/batches/{id}DELETE /api/v1/batches/{id}POST /api/v1/batches/{id}/cancelDELETE /api/v1/batches/delete-completedPOST /api/v1/classifyGET /api/v1/combosPOST /api/v1/completionsGET /api/v1/filesPOST /api/v1/filesGET /api/v1/files/{id}DELETE /api/v1/files/{id}GET /api/v1/files/{id}/contentPOST /api/v1/images/editsGET /api/v1/images/upscalePOST /api/v1/images/upscaleGET /api/v1/media/{id}POST /api/v1/issues/reportGET /api/v1/management/proxiesPOST /api/v1/management/proxiesPATCH /api/v1/management/proxiesDELETE /api/v1/management/proxiesGET /api/v1/management/proxies/assignmentsPUT /api/v1/management/proxies/assignmentsPUT /api/v1/management/proxies/bulk-assignGET /api/v1/management/proxies/healthGET /api/v1/me/statusGET /api/v1/muse-code/modelsGET /api/v1/music/generationsPOST /api/v1/music/generationsGET /api/v1/providers/{provider}/limitsPUT /api/v1/providers/{provider}/limitsGET /api/v1/quotas/checkGET /api/v1/registered-keysPOST /api/v1/registered-keysGET /api/v1/registered-keys/{id}DELETE /api/v1/registered-keys/{id}POST /api/v1/registered-keys/{id}/revokePOST /api/v1/relay/chat/completionsPOST /api/v1/relay/chat/completions/bifrostPOST /api/v1/responses/{path}GET /api/v1/search/analyticsPOST /api/v1/segmentGET /api/v1/video-bridge/drilldownDELETE /api/v1/video-bridge/drilldownGET /api/v1/videos/generationsPOST /api/v1/videos/generationsGET /api/v1/vscode/{token}POST /api/v1/vscode/{token}/api/chatPOST /api/v1/vscode/{token}/api/showGET /api/v1/vscode/{token}/api/tagsGET /api/v1/vscode/{token}/api/versionPOST /api/v1/vscode/{token}/chat/completionsGET /api/v1/vscode/{token}/combosGET /api/v1/vscode/{token}/modelsPOST /api/v1/vscode/{token}/responsesPOST /api/v1/vscode/{token}/v1/chat/completionsGET /api/v1/vscode/{token}/v1/modelsGET /api/v1/vscode/combos/{token}/{{slug}}POST /api/v1/vscode/combos/{token}/{{slug}}GET /api/v1/vscode/raw/{token}POST /api/v1/vscode/raw/{token}/api/chatPOST /api/v1/vscode/raw/{token}/api/showGET /api/v1/vscode/raw/{token}/api/tagsGET /api/v1/vscode/raw/{token}/api/versionPOST /api/v1/vscode/raw/{token}/chat/completionsGET /api/v1/vscode/raw/{token}/combosGET /api/v1/vscode/raw/{token}/modelsPOST /api/v1/vscode/raw/{token}/responsesPOST /api/v1/vscode/raw/{token}/v1/chat/completionsGET /api/v1/vscode/raw/{token}/v1/modelsPOST /api/v1/web/fetch
Payloads
See the full OpenAPI specification at GET /api/openapi/spec or docs/openapi.yaml for detailed request/response schemas.
Chat completions
Requires OMNIROUTE_URL and OMNIROUTE_KEY. See entry-point SKILL for setup.
Endpoints
POST $OMNIROUTE_URL/v1/chat/completions— OpenAI formatPOST $OMNIROUTE_URL/v1/messages— Anthropic Messages formatPOST $OMNIROUTE_URL/v1/responses— OpenAI Responses API
Discover
curl $OMNIROUTE_URL/v1/models | jq '.data[].id'
Combos (e.g. auto, cost-optimized, subscription) auto-fallback through multiple providers.
OpenAI format example
curl -X POST $OMNIROUTE_URL/v1/chat/completions \
-H "Authorization: Bearer $OMNIROUTE_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "claude-opus-4-7",
"messages": [{"role": "user", "content": "Refactor this function"}],
"stream": true
}'
Anthropic format example
curl -X POST $OMNIROUTE_URL/v1/messages \
-H "Authorization: Bearer $OMNIROUTE_KEY" \
-H "anthropic-version: 2023-06-01" \
-H "Content-Type: application/json" \
-d '{
"model": "claude-opus-4-7",
"max_tokens": 4096,
"messages": [{"role": "user", "content": "Hi"}]
}'
Tool use
Supports OpenAI tools array and Anthropic tools block. Tool results
auto-compressed via RTK (47 filters: git-diff, grep, test-jest, terraform-plan,
docker-logs, etc.) — 20-40% token savings. Disable per-request with
X-Omniroute-Rtk: off header.
Reasoning / thinking
Anthropic extended thinking and OpenAI Responses reasoning blocks are forwarded verbatim. Cached automatically via reasoning cache.
Errors
401→ invalid API key400 invalid_model→ model not in registry; check/v1/models503 circuit_open→ provider circuit breaker tripped; retry later or use combo429 rate_limited→ honorRetry-After; consider using a combo for auto-fallback
Image generation
Requires OMNIROUTE_URL and OMNIROUTE_KEY. See entry-point SKILL for setup.
Endpoints
POST $OMNIROUTE_URL/v1/images/generations— Text-to-imagePOST $OMNIROUTE_URL/v1/images/edits— Image edit (mask)POST $OMNIROUTE_URL/v1/images/variations— Variations
Discover
curl $OMNIROUTE_URL/v1/models/image | jq '.data[]'
Returns { id, owned_by, sizes:[...], capabilities:[...] } per model.
...
ユーザーレビュー (0)
レビューを書く
レビューなし
統計データ
ユーザー評価
この Skill を評価