Model marketplace
OpenSmartRoute vs OpenRouter
OpenRouter gives you hundreds of models behind one API with credits and a market-ranked auto mode. OpenSmartRoute decides across every capability you own, not just models, enforces your policy first and learns from your outcomes anywhere you run it.
“OpenRouter for every capability, not just models - and you can run it yourself.”
Choose OpenRouter when
- You only have models and want credits with zero ops
- Provider breadth is the purchase
Choose OpenSmartRoute when
- Agents, tools, workflows or human queues are in the routing pool
- PII, region or tenant policy must hold before any call
- Learning must be on your outcomes and your tenants
- Self-hosting or air-gap is required
Capability by capability
Our reading of public documentation; corrections welcome via GitHub issues.
| Capability | OpenSmartRoute | OpenRouter |
|---|---|---|
| Routes to LLMs | Yes | Yes |
| Routes to agents, skills, tools, workflows and humans | Yes | No |
| Hard policy before scoring (region, data boundary, PII, tenant, cost) | Yes | No |
| Per-request trace with ranked candidates and rejections | Yes | No |
| Learns from your outcomes (per target, per tenant)Market-wide ranking, not your tenants' outcomes | Yes | Partial |
| Multi-step plans (persona -> skill -> model) | Yes | No |
| Self-hosted, air-gapped, zero runtime dependencies | Yes | No |
| Prompt-injection and steering-gadget defences on the router | Yes | No |
| Savings ledger and Routing Audit | Yes | No |
| OpenAI-compatible proxy | Yes | Yes |
| Provider marketplace with credits | No | Yes |
| Request/response observability UI for every callDecision traces and usage; pair with your APM for payload-level tracing | Partial | Partial |
Use OpenRouter as a target
Keep OpenRouter for model breadth and let OpenSmartRoute decide when a request should go there at all versus to your on-prem model, an agent or a person.
- id: openrouter-frontier
kind: llm
description: Frontier model via OpenRouter
capabilities: { domains: [general, coding], max_complexity: 1.0 }
constraints: { data_boundary: public, pii_allowed: false }
cost: { usd_per_1k_tokens: 0.009 }Prove it on your traffic
Run the Routing Audit on a week of anonymised logs: savings versus your current mix, policy violations your routing allowed, and where agents or humans would have been the better answer.
Other comparisons: LiteLLM, Portkey, Building it yourself.