Provider Fallback Outage Routing
OpenRouter · OpenRouter
LLM routing and aggregation — OpenRouter
Evaluates OpenRouter's Provider Fallback & Outage Routing across 10 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's LLM routing and aggregation eval coverage.
About OpenRouter
OpenRouter is a unified LLM routing layer that gives developers access to hundreds of models through a single OpenAI-compatible API. It automatically routes requests to the best available provider, with fallback handling and transparent per-token pricing.
Sample tests· showing 3 of 10
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | Production gateway sends model anthropic/claude-3.5-sonnet with default provider routing; primary Anthropic endpoint times out with 502; secondary providers remain healthy per GET /api/v1/models. | Router retries on an alternate eligible provider for the same model slug and only returns 502 after exhausting documented provider-level retries, preserving the OpenAI-compatible error envelope. | Pass / FailRoutingcritical |
| 02 | Compliance workload pins provider.order to a single EU host; allow_fallbacks explicitly false in request body to POST /api/v1/chat/completions. | No silent reroute to other providers; client receives failure once the pinned provider is exhausted, with error body indicating routing could not satisfy constraints. | Pass / FailRoutinghigh |
| 03 | Request sets models: [openai/gpt-4o, anthropic/claude-3-haiku] with route fallback enabled; gpt-4o responds 200 but at extreme latency; engineer wants model-level degradation not provider-only retries. | Explains model-level fallback triggers per OpenRouter docs (including route: fallback) distinct from provider retries on 5xx/429, and when latency alone does not trigger model switch. | Pass / FailRoutinghigh |
How this eval is graded
Grade against expected.ideal_behavior and expected.rubric.
Rubric criteria
- Openrouter
- Ai Platform
- Provider Fallback Outage Routing
Recommended for
Works with
Related evals
Claude API
Evaluates Anthropic's Batch API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Extended Thinking across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Files API & Citations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
ViewFrequently asked questions
What does the Provider Fallback Outage Routing eval for OpenRouter OpenRouter test?+
Evaluates OpenRouter's Provider Fallback & Outage Routing across 10 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's LLM routing and aggregation eval coverage.
How is the Provider Fallback Outage Routing eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric.
How many test cases does this eval pack include?+
The Provider Fallback Outage Routing pack for OpenRouter OpenRouter contains 10 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Provider Fallback Outage Routing pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.