Back to Details

omniroute

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

OmniRoute

OmniRoute

Local/remote AI gateway exposing OpenAI-compatible REST. One key, 207+ providers, auto-fallback, RTK token saver, MCP server, A2A agents.
支持 OpenAI 兼容 REST 接口的本地/远程 AI 网关。一个密钥接入 207+ 服务商，支持自动回退、RTK token 节省器、MCP server 和 A2A agent。

Setup

配置

bash
export OMNIROUTE_URL="http://localhost:20128"      # or VPS / tunnel URL
export OMNIROUTE_KEY="sk-..."                       # from Dashboard → API Keys
All requests: 
${OMNIROUTE_URL}/v1/...
with 
Authorization: Bearer ${OMNIROUTE_KEY}
.
Verify: 
curl $OMNIROUTE_URL/api/health
→ 
{"ok":true}
bash
export OMNIROUTE_URL="http://localhost:20128"      # 或使用 VPS / 隧道 URL
export OMNIROUTE_KEY="sk-..."                       # 来自 Dashboard → API Keys
所有请求：
${OMNIROUTE_URL}/v1/...
，并设置 
Authorization: Bearer ${OMNIROUTE_KEY}
。
验证：
curl $OMNIROUTE_URL/api/health
→ 
{"ok":true}

Discover models

发现模型

bash
curl $OMNIROUTE_URL/v1/models                  # chat/LLM (default)
curl $OMNIROUTE_URL/v1/models/image            # image-gen
curl $OMNIROUTE_URL/v1/models/tts              # text-to-speech
curl $OMNIROUTE_URL/v1/models/embedding        # embeddings
curl $OMNIROUTE_URL/v1/models/web              # web search + fetch
curl $OMNIROUTE_URL/v1/models/stt              # speech-to-text
Use 
data[].id
as 
model
field in requests. Combos appear with 
owned_by:"combo"
.
bash
curl $OMNIROUTE_URL/v1/models                  # 聊天/LLM（默认）
curl $OMNIROUTE_URL/v1/models/image            # 图像生成
curl $OMNIROUTE_URL/v1/models/tts              # 文本转语音
curl $OMNIROUTE_URL/v1/models/embedding        # embeddings
curl $OMNIROUTE_URL/v1/models/web              # 网页搜索 + 抓取
curl $OMNIROUTE_URL/v1/models/stt              # 语音转文本
在请求中将 
data[].id
用作 
model
字段。组合模型会显示 
owned_by:"combo"
。

Capability skills

能力 skills

CapabilityRaw URL
Chat / code-genhttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-chat/SKILL.md
Image generationhttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-image/SKILL.md
Text-to-speechhttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-tts/SKILL.md
Speech-to-texthttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-stt/SKILL.md
Embeddingshttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-embeddings/SKILL.md
Web searchhttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-web-search/SKILL.md
Web fetchhttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-web-fetch/SKILL.md
MCP server (37 tools)https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-mcp/SKILL.md
A2A protocolhttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-a2a/SKILL.md
Routing & comboshttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-routing/SKILL.md
Token compressionhttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-compression/SKILL.md
Monitoring & healthhttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-monitoring/SKILL.md
能力原始 URL
聊天 / 代码生成https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-chat/SKILL.md
图像生成https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-image/SKILL.md
文本转语音https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-tts/SKILL.md
语音转文本https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-stt/SKILL.md
Embeddingshttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-embeddings/SKILL.md
网页搜索https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-web-search/SKILL.md
网页抓取https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-web-fetch/SKILL.md
MCP server（37 个工具）https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-mcp/SKILL.md
A2A 协议https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-a2a/SKILL.md
路由与组合https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-routing/SKILL.md
Token 压缩https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-compression/SKILL.md
监控与健康检查https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-monitoring/SKILL.md

CLI skills (omniroute binary)

CLI skills（omniroute binary）

CapabilityRaw URL
CLI entry pointhttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli/SKILL.md
CLI admin & lifecyclehttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-admin/SKILL.md
CLI providers & keyshttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-providers/SKILL.md
CLI cloud agentshttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-cloud/SKILL.md
CLI evals & benchmarkshttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-eval/SKILL.md
能力原始 URL
CLI 入口https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli/SKILL.md
CLI 管理与生命周期https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-admin/SKILL.md
CLI 服务商与密钥https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-providers/SKILL.md
CLI 云端 agenthttps://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-cloud/SKILL.md
CLI evals 与基准测试https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute-cli-eval/SKILL.md

Errors

错误

  • 401
    → set/refresh 
    OMNIROUTE_KEY
    (Dashboard → API Keys)
  • 400 Invalid model format
    → check 
    model
    exists in 
    /v1/models/<kind>
  • 503 Provider circuit open
    → upstream provider down; retry after 
    Retry-After
    seconds
  • 429
    → rate limited; honor 
    Retry-After
  • 401
    → 设置或刷新 
    OMNIROUTE_KEY
    （Dashboard → API Keys）
  • 400 Invalid model format
    → 检查 
    model
    是否存在于 
    /v1/models/<kind>
    中
  • 503 Provider circuit open
    → 上游提供商不可用；在 
    Retry-After
    指定的秒数后重试
  • 429
    → 触发速率限制；遵循 
    Retry-After

Differentiators vs OpenAI direct

相比直接使用 OpenAI 的差异化优势

  • Auto-fallback combos (14 strategies): never stop coding even if a provider rate-limits
  • RTK token saver: tool_result compressed via 47 specialized filters (git-diff, test-jest, terraform-plan, docker-logs…) — 20-40% token reduction
  • Caveman mode: optional terse system prompt injection (LITE/FULL/ULTRA) — 15-25% completion reduction
  • MCP + A2A servers built-in (this is the only AI router that exposes both protocols)
  • Memory with FTS5 + Qdrant for persistent agent context
  • Guardrails for PII masking, prompt injection detection, vision policies
  • 自动回退组合（14 种策略）：即使提供商触发速率限制，也不会停止编码
  • RTK token 节省器：通过 47 个专用过滤器压缩 tool_result（git-diff、test-jest、terraform-plan、docker-logs……）——减少 20–40% token
  • Caveman 模式：可选的简洁系统提示词注入（LITE/FULL/ULTRA）——减少 15–25% 的补全内容
  • 内置 MCP + A2A server（这是唯一同时暴露两种协议的 AI 路由器）
  • Memory：使用 FTS5 + Qdrant 实现持久化 agent 上下文
  • Guardrails：支持 PII 掩码、提示词注入检测和视觉策略