Current ZDR-capable OpenRouter endpoints rather than assuming privacy support from the model name alone.
Zero Data Retention model selection
Require ZDR without throwing away the rest of the decision.
Privacy can be a hard constraint, but it is not the whole model-selection problem. ModelShortlist can filter to current OpenRouter ZDR-capable endpoints while still checking context, tools, pricing, and workload fit.
The decision
The answer changes with the workload.
A privacy requirement should narrow the candidate set, not replace the rest of your reasoning. When you explicitly require ZDR, ModelShortlist checks the current endpoint-level data and keeps hard constraints tied to the same eligible endpoint. When you do not require ZDR, it does not silently exclude the broader OpenRouter catalog.
ModelShortlist identifies current ZDR-capable options for selection. When you later send inference requests through OpenRouter, enforce ZDR again in the actual provider request rather than treating model selection alone as the privacy control.
Tool and parameter support on the same endpoint that satisfies the ZDR requirement.
Minimum context and pricing constraints after the privacy filter is applied.
Latency, throughput, uptime, and provider options when endpoint-level operational data is relevant.
Independent performance evidence
Artificial Analysis
ModelShortlist uses Artificial Analysis benchmark and performance evidence when the model identity can be reconciled confidently. It does not create, relabel, or pretend ownership of those benchmarks.
How Artificial Analysis contributesCurrent operational facts
OpenRouter
The current catalog supplies model availability, context, supported parameters, pricing, and provider details. ZDR endpoint evidence is applied only when the workload explicitly requires ZDR.
Compare OpenRouter models by workloadModel selection is time-sensitive: new models launch, prices move, benchmark results change, tool support evolves, and providers add or remove endpoints. ModelShortlist surfaces freshness and degraded upstream state instead of silently presenting stale evidence as current.
Ask naturally
Prompts that work.
ModelShortlist sits behind the AI assistant you already use. Describe the job and hard constraints instead of translating them into a fixed ranking formula.
“I need 200k context, tool calling, and ZDR. What are my best current options?”
“ZDR is mandatory for this workload. Keep output cost under $10 per million tokens and prioritize strong reasoning.”
“Compare my best ZDR-capable options for an autonomous coding agent with at least 100k context.”
Why ModelShortlist
Current evidence, not a static leaderboard.
ZDR is an explicit optional constraint, not a default filter that distorts every recommendation.
When ZDR is required, endpoint-level eligibility and other hard constraints are checked together.
OpenRouter capability and operational data can be combined with independent Artificial Analysis benchmarks where a confident match exists.
The host AI can explain what quality, cost, context, or provider tradeoffs remain after the privacy requirement is satisfied.
Local BYOK architecture. No ModelShortlist account, hosted key vault, or telemetry in the MCP.
Put current model-selection evidence inside your assistant.
Install the local MCP, add your OpenRouter and Artificial Analysis keys, and ask the model-selection question in normal language.
Open the install configurator