Core control
Keep requests moving
When a target is unavailable or out of quota, OmniRoute selects the next healthy target.
OmniRoute / Product capabilities
Controls you can inspect and self-host.
Review the gateway controls, platform services and local dashboard without treating a feature list as a hosting promise.
Core controls
What ships with the self-hosted gateway.
Core control
Compress eligible context
A 12-engine pipeline reduces eligible context by 15–95%. The documented tool-heavy example reduced it by ~89%.
Core control
$0 to start
90+ free tiers and 56 recurring or keyless free-forever providers. No credit card needed for the keyless quick start.
Core control
35 supported integrations
Configure Claude Code, Codex, Cursor, Cline, Copilot, Antigravity and other supported clients through one endpoint.
Core control
One endpoint
Accept OpenAI, Claude, Gemini and Responses API requests at /v1.
Core control
Operational controls
Circuit breakers, encrypted credentials, MCP, A2A, memory, guardrails and evals ship with the self-hosted gateway.
Capabilities
Configure the request path from one dashboard.
Configure fallback, quota, compression, memory, MCP and A2A from the local dashboard.
Gateway control
Smart Combos
Set a model to auto or build your own combo with 19 strategies and tiered fallback.
Use a ready-made mode or combine 19 strategies around the goal that matters.
Gateway control
Memory
Persistent conversational memory with FTS5 keyword and Qdrant vector recall.
Gateway control
3-layer resilience
Provider breaker, connection cooldown and model lockout isolate failures at the smallest useful scope.
Gateway control
Free quota
Pool-deduped accounting across 90+ free tiers.
Published free-tier capacity, deduplicated.
Gateway control
Compression
12 composable engines for tool output, context and response style.
Reduce eligible tokens by 15–95%.
Gateway control
MCP server
A built-in server exposing 110 gateway tools across 33 scopes.
110 tools · 33 scopes · stdio / HTTP
Gateway control
A2A protocol
A JSON-RPC agent server with 6 skills.
6 skills · JSON-RPC 2.0 · Agent Card
Gateway control
Semantic cache
A repeated request can return from cache, and the response header reports what that saved.
X-OmniRoute-Cache: HIT
Platform services
Extend the local gateway.
Embedded services
Run Bifrost, 9Router or CLIProxy as managed sidecars, or operate a full cluster profile.
Your knowledge, in context
Connect Notion and Obsidian as MCP-backed context sources.
Guardrails built in
Prompt-injection checks, opt-in PII redaction and normalized safety filters without provider lock-in.
Usage goals
Track streaks, achievements and live savings alongside operational analytics.
Dashboard
Inspect configuration and example operating views.
The values below are example data. Your local dashboard reflects your own providers, requests and costs.
352providers connected
90+free tiers available
198healthy in this example
- Claude Code · OAuth · live
- Cerebras · free · live
- Groq · cooldown example
- NVIDIA NIM · free · live
19routing strategies
8msexample failover
- auto · 15-factor scoring
- auto/cheap · cost optimized
- auto/coding · quality first
1.42Mrequests · 30d example
99.4%success rate example
312msp50 latency example
- Requests over time
- Provider share
- Model and endpoint usage
3resilience layers
2keys cooling down in example
- Provider circuit breakers
- Connection cooldowns
- Model lockouts
$0.00spent · 30d example
1.51Bfree tokens routed in example
- Free tiers
- Flat-fee subscriptions
- Paid APIs