{"provider":{"title":"Ask Model (Session)","description":"Query 1-5 frontier AI models in parallel via OpenRouter. Session-priced: open a channel once, run many inferences against it, close when done. Per-voucher advance: $0.02. Suggested deposit: $0.05. Models: openai/gpt-4.1, openai/gpt-4.1-mini, openai/gpt-4.1-nano, anthropic/claude-opus-4-6, anthropic/claude-sonnet-4-5, anthropic/claude-haiku-4-5, google/gemini-2.5-pro, google/gemini-2.5-flash, meta-llama/llama-3.3-70b-instruct, mistralai/mistral-large, mistralai/mistral-small, deepseek/deepseek-v3, deepseek/deepseek-r1. Returns a list of per-model responses with token counts, latency, and estimated raw cost. Note: per-voucher pricing is flat; for true per-token metered billing wait for Phase 3 metered intent. Surface map: session-paid `POST /run` (one voucher per multi-model call); free `GET /models` (per-million token costs + estimated per-call cost per model — useful before opening a channel).","paidRoutes":[{"method":"POST","path":"/run","protocol":"mpp","sku":"ask-model","priceUsdcMicro":"20000","pricingVersion":1,"description":"Run a prompt against 1-5 models in parallel. Per-voucher: $0.02. Buyer opens a session channel (recommended deposit $0.05) and signs cumulative vouchers per call. Each call advances the channel by $0.02. Close the session to get unspent deposit refunded. Pre-flight: free `GET /models` lists every supported model with per-million token costs and an estimated per-call cost — use it to choose the `models` array before opening the channel.","intent":"session","suggestedDepositUsdcMicro":"50000","unitType":"request"}],"freeRoutes":[{"method":"GET","path":"/models","description":"List supported models with per-million-token input/output costs and an estimated per-call cost (assumes ~500/500 input/output tokens with the 50% margin applied)."},{"method":"GET","path":"/"},{"method":"GET","path":"/contract"},{"method":"GET","path":"/openapi.json"},{"method":"GET","path":"/health"}]}}