Current OpenRouter input and output pricing, not remembered launch pricing.
Cost-efficient tool-calling models
Lower model cost without breaking the agent loop.
The cheapest token price is not useful if the model cannot reliably use the tools your workflow depends on. ModelShortlist lets cost become a constraint while preserving capabilities and workload fit.
The decision
The answer changes with the workload.
High-volume agents, extraction pipelines, and repetitive subagents can turn small per-token differences into large bills. But a low price only creates value when the candidate still has the context, tool support, and quality level needed to finish the work without retries or escalation.
Tool/function-calling support as a hard eligibility requirement when the workflow depends on it.
Context limits large enough for the task so cheap models are not selected into an impossible workload.
Quality and performance evidence that helps distinguish inexpensive from merely underpowered.
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.
“What is the cheapest model I would trust for repetitive tool-using coding subagents?”
“Find my best tool-calling options under $5 per million output tokens with at least 64k context.”
“I have a high-volume extraction workflow that uses tools. Minimize cost without dropping below a reasonable quality bar.”
Why ModelShortlist
Current evidence, not a static leaderboard.
Price ceilings can be applied before the host AI reasons about the remaining candidates.
Tool use and context can remain hard requirements instead of becoming secondary notes on a cheap-model list.
Artificial Analysis metrics are attached when confidently matched, giving the host AI additional evidence about the cost/quality tradeoff.
The result can distinguish a lowest-cost candidate from a best-value candidate when paying slightly more materially improves fit.
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