{"object":"service","service":"inference-gateway","version":"0.1.0","description":"Payment-gated inference gateway for Routify. Clients can discover the model catalog, inspect x402 payment rules, and submit OpenAI-compatible chat completion requests.","timestamp":"2026-09-26T21:50:52.848Z","upstream":{"provider":"OpenRouter","baseUrl":"https://openrouter.ai/api/v1","overrideEnabled":true,"resolvedOverrideModel":"nvidia/nemotron-3-ultra-550b-a55b:free","routingNote":"Development override is enabled. Every logical model currently resolves to the same upstream model even though pricing and logical selection still follow the requested model id."},"payment":{"protocol":"x402-v2","network":"testnet","caip2":"algorand:SGO1GKSzyE7IEPItTxCByw9x8FmnrCDexi9/cOUJOiI=","facilitatorUrl":"https://facilitator.goplausible.xyz","payTo":"TLCDFH7VII4HIOD2DNGPOVEMAVJIZC4BJQKYA27SDSDILJLVJUGHNCIU6U","asset":{"id":"10458941","symbol":"USDC","decimals":6},"pricingStrategy":"Price is determined from the requested logical model id before the upstream model call is made.","challengeTag":"x402-global-challenge","bazaarDiscovery":true,"bypass":{"enabled":false}},"endpoints":[{"method":"GET","path":"/","description":"Service discovery manifest for machine and frontend clients."},{"method":"GET","path":"/health","description":"Basic runtime health, routing mode, and x402 network configuration."},{"method":"GET","path":"/v1/models","description":"Logical model catalog with routing metadata, pricing, and selection guidance."},{"method":"POST","path":"/v1/chat/completions","description":"Paid OpenAI-compatible chat completions endpoint. Accepts a chat payload, charges through Algorand x402, then forwards to OpenRouter.","contentType":"application/json","paymentRequired":true}],"requestContract":{"route":"POST /v1/chat/completions","bodyFormat":"OpenAI-compatible chat completion payload","requiredFields":["messages"],"optionalFields":["model","stream","max_tokens","max_completion_tokens","temperature","top_p","response_format","tools","tool_choice","metadata","user"],"defaults":{"model":"auto","stream":false},"modelSelection":{"automatic":"If model is omitted, the gateway uses the logical model id 'auto'.","explicit":"If model is provided, it must match one of the ids from GET /v1/models."},"example":{"model":"auto","messages":[{"role":"user","content":"Write a short paragraph about why Algorand is good for micropayments."}],"stream":false}},"responseContract":{"successModes":["JSON","streaming"],"responseHeaders":[{"name":"x-routify-requested-model","description":"Logical model id requested by the client, or auto when omitted."},{"name":"x-routify-intended-model","description":"Production upstream model slug associated with the logical model."},{"name":"x-routify-resolved-model","description":"Actual OpenRouter model slug used for the upstream request."},{"name":"x-routify-tier","description":"Logical pricing and quality tier for the selected model."},{"name":"x-routify-price-usdc","description":"USDC amount charged for the selected logical model."}]},"mainnetReady":{"networkSwitch":"Set X402_NETWORK=mainnet; USDC changes automatically to ASA 31566704.","testModeAllowed":true},"models":[{"id":"auto","provider":"Routify","displayName":"Automatic","description":"Uses the gateway default routing mode when the user does not explicitly pick a model.","recommendedFor":["General prompts","Users who do not want to choose manually","Default app flows"],"selectionHint":"Use this when the user does not care about provider choice and wants the gateway default.","capabilities":["Planning","Writing","Research","Tools"],"speed":"balanced","cost":"medium","availability":"available","tier":"balanced","intendedUpstreamModel":"openrouter/auto","priceUsd":"$0.010","priceUsdc":"0.010","resolvedUpstreamModel":"nvidia/nemotron-3-ultra-550b-a55b:free"},{"id":"openai/gpt-5.6-luna","provider":"OpenAI","displayName":"GPT-5.6 Luna","description":"Fast OpenAI option for short writing, summaries, and lighter prompt work.","recommendedFor":["Short prompts","Summaries","Fast UI interactions"],"selectionHint":"Pick this when latency matters more than depth.","capabilities":["Writing","Chat","Reasoning"],"speed":"fast","cost":"low","availability":"available","tier":"fast","intendedUpstreamModel":"openai/gpt-5.6-luna","priceUsd":"$0.006","priceUsdc":"0.006","resolvedUpstreamModel":"nvidia/nemotron-3-ultra-550b-a55b:free"},{"id":"openai/gpt-5.6-terra","provider":"OpenAI","displayName":"GPT-5.6 Terra","description":"Balanced OpenAI option for everyday reasoning, coding, and agent tasks.","recommendedFor":["General reasoning","Coding","Agent flows"],"selectionHint":"Pick this as the balanced OpenAI default.","capabilities":["Writing","Reasoning","Coding","Tools"],"speed":"balanced","cost":"medium","availability":"available","tier":"balanced","intendedUpstreamModel":"openai/gpt-5.6-terra","priceUsd":"$0.015","priceUsdc":"0.015","resolvedUpstreamModel":"nvidia/nemotron-3-ultra-550b-a55b:free"},{"id":"openai/gpt-5.6-sol","provider":"OpenAI","displayName":"GPT-5.6 Sol","description":"Premium OpenAI option for harder reasoning, coding, and long-horizon tasks.","recommendedFor":["Hard reasoning","Complex coding","Long multi-step tasks"],"selectionHint":"Pick this when quality matters more than price and speed.","capabilities":["Writing","Reasoning","Coding","Tools"],"speed":"slow","cost":"high","availability":"available","tier":"premium","intendedUpstreamModel":"openai/gpt-5.6-sol","priceUsd":"$0.030","priceUsdc":"0.030","resolvedUpstreamModel":"nvidia/nemotron-3-ultra-550b-a55b:free"},{"id":"anthropic/claude-haiku-4.5","provider":"Anthropic","displayName":"Claude Haiku 4.5","description":"Fast Claude option with strong responsiveness for high-volume work.","recommendedFor":["Fast drafting","High request volume","Simple assistants"],"selectionHint":"Pick this for cheaper fast Claude-style responses.","capabilities":["Writing","Reasoning","Tools"],"speed":"fast","cost":"low","availability":"available","tier":"fast","intendedUpstreamModel":"anthropic/claude-haiku-4.5","priceUsd":"$0.006","priceUsdc":"0.006","resolvedUpstreamModel":"nvidia/nemotron-3-ultra-550b-a55b:free"},{"id":"anthropic/claude-sonnet-4.6","provider":"Anthropic","displayName":"Claude Sonnet 4.6","description":"Balanced Claude option for strong coding, document work, and agents.","recommendedFor":["Coding","Document work","Balanced agent tasks"],"selectionHint":"Pick this as the balanced Claude default.","capabilities":["Writing","Reasoning","Coding","Tools"],"speed":"balanced","cost":"medium","availability":"available","tier":"balanced","intendedUpstreamModel":"anthropic/claude-sonnet-4.6","priceUsd":"$0.015","priceUsdc":"0.015","resolvedUpstreamModel":"nvidia/nemotron-3-ultra-550b-a55b:free"},{"id":"anthropic/claude-opus-4.7","provider":"Anthropic","displayName":"Claude Opus 4.7","description":"Premium Claude option for persistent, complex, multi-step work.","recommendedFor":["Deep analysis","Complex workflows","Long task chains"],"selectionHint":"Pick this for the strongest Claude-tier reasoning.","capabilities":["Writing","Reasoning","Coding","Tools"],"speed":"slow","cost":"high","availability":"available","tier":"premium","intendedUpstreamModel":"anthropic/claude-opus-4.7","priceUsd":"$0.030","priceUsdc":"0.030","resolvedUpstreamModel":"nvidia/nemotron-3-ultra-550b-a55b:free"},{"id":"meta-llama/llama-3.3-70b-instruct","provider":"Meta","displayName":"Llama 3.3 70B Instruct","description":"Strong open-source general-purpose text model.","recommendedFor":["General chat","Budget-conscious prompts","Open-source preference"],"selectionHint":"Pick this when you want a strong open-source default.","capabilities":["Writing","Reasoning","Chat"],"speed":"balanced","cost":"low","availability":"available","tier":"open-source","intendedUpstreamModel":"meta-llama/llama-3.3-70b-instruct","priceUsd":"$0.003","priceUsdc":"0.003","resolvedUpstreamModel":"nvidia/nemotron-3-ultra-550b-a55b:free"},{"id":"mistralai/mistral-small-24b-instruct-2501","provider":"Mistral","displayName":"Mistral Small 3","description":"Low-cost open-source model for lightweight text generation and chat.","recommendedFor":["Lightweight prompts","Simple chat","Lowest-cost explicit pick"],"selectionHint":"Pick this for cheap lightweight prompt handling.","capabilities":["Writing","Chat"],"speed":"fast","cost":"low","availability":"available","tier":"open-source","intendedUpstreamModel":"mistralai/mistral-small-24b-instruct-2501","priceUsd":"$0.003","priceUsdc":"0.003","resolvedUpstreamModel":"nvidia/nemotron-3-ultra-550b-a55b:free"},{"id":"google/gemma-3-27b-it","provider":"Google","displayName":"Gemma 3 27B","description":"Open-source Google model for general reasoning and multimodal-friendly workflows.","recommendedFor":["General reasoning","Open-source Google-style flows","Tool-friendly tasks"],"selectionHint":"Pick this when you want open-source reasoning with a Google-family slot.","capabilities":["Writing","Reasoning","Tools"],"speed":"balanced","cost":"low","availability":"available","tier":"open-source","intendedUpstreamModel":"google/gemma-3-27b-it","priceUsd":"$0.003","priceUsdc":"0.003","resolvedUpstreamModel":"nvidia/nemotron-3-ultra-550b-a55b:free"},{"id":"google/gemini-2.5-flash","provider":"Google","displayName":"Gemini 2.5 Flash","description":"Fast practical multimodal workhorse for general prompts and tool-heavy tasks.","recommendedFor":["Fast general use","Tool-heavy flows","Everyday prompts"],"selectionHint":"Pick this for a fast Google-family general model.","capabilities":["Writing","Reasoning","Coding","Tools"],"speed":"fast","cost":"low","availability":"available","tier":"fast","intendedUpstreamModel":"google/gemini-2.5-flash","priceUsd":"$0.006","priceUsdc":"0.006","resolvedUpstreamModel":"nvidia/nemotron-3-ultra-550b-a55b:free"}]}