1gateway AI API
The Ai API from 1gateway — 1 operation(s) for ai.
Work with this as data
Every API here is available over the APIs.io API and to AI agents over MCP.
MCP server
One button, every client — Claude, Cursor, VS Code and the rest.
https://apis.io/mcp
Tools for apis
7 MCP tools reach this
find_apisBrowse and filter every API in the catalog.get_api_artifactsOne API's artifacts, grouped by type.get_openapiThe primary OpenAPI for this API.find_similar_apisAPIs that look like this one.apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.resolveTurn a domain, URL or GitHub org into the provider it belongs to.find_cohortsEvery scored population of providers in the catalog.
Call it yourself
curl for this page
This API
curl "https://apis.io/api/v1/apis/1gateway-dev:1gateway-dev-ai-api"
All apis
curl "https://apis.io/api/v1/apis?limit=25"Calls without X-API-Key need no key. Calls with it need a paid plan — get a key.
Get an API key
Free tier, no form to fill in. Signing in shares your email address with us — we store it to create your key and to recognise you if you sign in with another provider. See our Privacy Policy and Terms.
A second provider on the same verified email joins the account you already have.
OpenAPI Specification
openapi: 3.2.0
info:
title: 1gateway AI API
version: 0.1.1
description: '1gateway lets AI agents buy the data and APIs they need, one call at a time: no signup, no API key, paid per request with USDC on Base or Solana via x402. Coverage: company and compliance checks (sanctions, KYB, SEC filings, Korean companies), Korean real estate, trade and macro data, onchain and market data, and agent utilities such as document-to-Markdown. Every response names its source. Failed calls are never charged.'
x-guidance: 'Use this API when an agent needs data as English JSON with pay-per-call access and no signup. Common tasks: Korea apartment sale transactions and district price summaries, Korean public holidays, Korea company KYB (company resolve, profile, NTS business status, DART filings, supplier check), Korean HWP/HWPX to Markdown, web page to Markdown (web read), and document extract (PDF, DOCX, HWP/HWPX or web page to structured JSON). Current coverage is South Korea: apartment sale prices (MOLIT), exports/imports and 10-day export flash (Korea Customs), Bank of Korea statistics (ECOS), weather forecast by lat/lon (KMA), holidays (KASI); more APIs and countries are coming. Base onchain reads: token risk, wallet balances, token info, transaction status. Also: document → Markdown conversion for web pages, PDF, DOCX and Korean HWP/HWPX (price by detected type, decided before payment). Every /v1 endpoint is paid: call it, receive HTTP 402 with PAYMENT-REQUIRED, pay with an x402 client (USDC on Base or Solana, same price) and retry. No charge on failure: settlement happens only on 2xx; 4xx/5xx are never charged. Agent guide: https://1gateway.dev/agents.md. Status: https://1gateway.dev/status. MCP (Streamable HTTP) is available at https://1gateway.dev/mcp: initialize and tools/list are free, tools/call is paid via x402 _meta.'
contact:
email: contact@1gateway.dev
servers:
- url: https://1gateway.dev
tags:
- name: AI
paths:
/v1/ai/llm-prices:
get:
operationId: llm_model_prices
summary: LLM model price lookup, token cost calculator and cheapest-model finder
description: 'Use when an agent must price an LLM call or pick the cheapest model. Input: model (up to 5, comma list) plus optional input_tokens, output_tokens, requests; or no model plus mode, provider, min_context, capabilities (vision, function_calling...). Output: JSON USD per 1M input/output tokens, cached/batch prices, context window, provider, capabilities, cost estimate or cheapest ranking, as_of + source commit. Example: ?model=gpt-4o&input_tokens=12000&output_tokens=800'
x-payment-info:
price:
mode: fixed
currency: USD
amount: '0.004000'
protocols:
- x402: {}
x-x402:
price: $0.004
scheme: exact
network: eip155:8453
networks:
- eip155:8453
- solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp
asset: USDC
kind: single
source: LiteLLM model_prices_and_context_window.json (MIT License, Copyright (c) 2023 Berri AI), pinned commit bundled at build time; snapshot date and commit returned as as_of/source
parameters:
- name: model
in: query
required: false
description: Model name(s), up to 5 comma-separated (gpt-4o, claude-sonnet-4-5, gemini/gemini-2.5-flash). Also lists the same model at other providers. Omit to rank the cheapest models. Unknown → 400 with suggestions (not charged)
schema:
type: string
maxLength: 400
example: gpt-4o
- name: input_tokens
in: query
required: false
description: Input (prompt) tokens per request, for the cost estimate and ranking
schema:
type: integer
minimum: 0
maximum: 100000000
example: 12000
- name: output_tokens
in: query
required: false
description: Output (completion) tokens per request
schema:
type: integer
minimum: 0
maximum: 100000000
example: 800
- name: requests
in: query
required: false
description: Number of requests to multiply the estimate by (default 1)
schema:
type: integer
minimum: 1
maximum: 1000000000
example: 1000
- name: provider
in: query
required: false
description: 'LiteLLM provider id filter: openai, anthropic, gemini, vertex_ai, bedrock, azure, groq, deepseek, mistral...'
schema:
type: string
pattern: ^[A-Za-z0-9_.-]{1,40}$
example: openai
- name: mode
in: query
required: false
description: Model type for the cheapest search (default chat; chat includes responses-API models). Ignored when model is given
schema:
type: string
enum:
- chat
- embedding
- completion
- rerank
- image_generation
- audio_transcription
- audio_speech
- realtime
- moderation
example: chat
- name: min_context
in: query
required: false
description: Minimum context window (max input tokens)
schema:
type: integer
minimum: 1
maximum: 100000000
example: 100000
- name: capabilities
in: query
required: false
description: 'Comma-separated required capabilities: vision, function_calling, reasoning, prompt_caching, response_schema, pdf_input, audio_input, audio_output, web_search, computer_use, video_input'
schema:
type: string
maxLength: 200
example: vision
- name: limit
in: query
required: false
description: Max rows per model or in the ranking (default 10)
schema:
type: integer
minimum: 1
maximum: 25
example: 5
- name: include_deprecated
in: query
required: false
description: Include models past their deprecation date in the cheapest search (default false)
schema:
type: string
enum:
- 'true'
- 'false'
example: 'false'
responses:
'200':
description: OK (payment settled). Example is a trimmed real output.
content:
application/json:
schema:
type: object
properties:
as_of:
type: string
source:
type: object
properties:
name:
type: string
commit:
type: string
committed_at:
type: string
url:
type: string
license:
type: string
models_in_table:
type: integer
kind:
type: string
query:
type: object
properties:
model:
type:
- array
- 'null'
items:
type: string
input_tokens:
type:
- integer
- 'null'
output_tokens:
type:
- integer
- 'null'
requests:
type: integer
provider:
type:
- string
- 'null'
mode:
type: string
min_context:
type:
- integer
- 'null'
capabilities:
type: array
items:
type: string
limit:
type: integer
include_deprecated:
type: boolean
results:
type: array
items:
type: object
properties:
query:
type: string
exact_match:
type: boolean
listings_total:
type: integer
listings:
type: array
items:
type: object
properties:
model:
type: string
provider:
type: string
mode:
type: string
input_per_1m_usd:
type:
- number
- 'null'
output_per_1m_usd:
type:
- number
- 'null'
cached_input_per_1m_usd:
type:
- number
- 'null'
batch_input_per_1m_usd:
type:
- number
- 'null'
batch_output_per_1m_usd:
type:
- number
- 'null'
context_window:
type:
- integer
- 'null'
max_output_tokens:
type:
- integer
- 'null'
capabilities:
type: array
items:
type: string
long_context_tier:
type:
- object
- 'null'
properties:
above_input_tokens:
type: integer
input_per_1m_usd:
type:
- number
- 'null'
output_per_1m_usd:
type:
- number
- 'null'
deprecation_date:
type:
- string
- 'null'
deprecated:
type: boolean
estimate:
type:
- object
- 'null'
properties:
input_tokens:
type: integer
output_tokens:
type: integer
requests:
type: integer
input_cost_usd:
type: number
output_cost_usd:
type: number
cost_per_request_usd:
type: number
total_cost_usd:
type: number
long_context_tier_applied:
type: boolean
attribution:
type: string
ranking:
type: string
matched_total:
type: integer
models:
type: array
items:
type: object
properties:
model:
type: string
provider:
type: string
mode:
type: string
input_per_1m_usd:
type:
- number
- 'null'
output_per_1m_usd:
type:
- number
- 'null'
cached_input_per_1m_usd:
type:
- number
- 'null'
batch_input_per_1m_usd:
type:
- number
- 'null'
batch_output_per_1m_usd:
type:
- number
- 'null'
context_window:
type:
- integer
- 'null'
max_output_tokens:
type:
- integer
- 'null'
capabilities:
type: array
items:
type: string
long_context_tier:
type:
- object
- 'null'
properties:
above_input_tokens:
type: integer
input_per_1m_usd:
type:
- number
- 'null'
output_per_1m_usd:
type:
- number
- 'null'
deprecation_date:
type:
- string
- 'null'
deprecated:
type: boolean
estimate:
type:
- object
- 'null'
properties:
input_tokens:
type: integer
output_tokens:
type: integer
requests:
type: integer
input_cost_usd:
type: number
output_cost_usd:
type: number
cost_per_request_usd:
type: number
total_cost_usd:
type: number
long_context_tier_applied:
type: boolean
required:
- as_of
- source
- kind
- query
- attribution
example:
as_of: '2026-10-18'
source:
name: LiteLLM model_prices_and_context_window.json
commit: af194c842cd8c0276f69fc05efacaa18505c7484
committed_at: '2026-10-18T02:54:27Z'
url: https://github.com/BerriAI/litellm/blob/af194c842cd8c0276f69fc05efacaa18505c7484/model_prices_and_context_window.json
license: MIT
models_in_table: 3418
kind: lookup
query:
model:
- gpt-4o
input_tokens: 12000
output_tokens: 800
requests: 1000
provider: null
mode: chat
min_context: null
capabilities: []
limit: 3
include_deprecated: false
results:
- query: gpt-4o
exact_match: true
listings_total: 5
listings:
- model: gpt-4o
provider: openai
mode: chat
input_per_1m_usd: 2.5
output_per_1m_usd: 10
cached_input_per_1m_usd: 1.25
batch_input_per_1m_usd: 1.25
batch_output_per_1m_usd: 5
context_window: 128000
max_output_tokens: 16384
capabilities:
- vision
long_context_tier: null
deprecation_date: null
deprecated: false
estimate:
input_tokens: 12000
output_tokens: 800
requests: 1000
input_cost_usd: 30
output_cost_usd: 8
cost_per_request_usd: 0.038
total_cost_usd: 38
long_context_tier_applied: false
attribution: 'Prices: LiteLLM model_prices_and_context_window.json, MIT License, Copyright (c) 2023 Berri AI. List prices as compiled by LiteLLM; confirm with the provider b…'
'400':
description: Invalid or missing parameters, or URL/domain not allowed (validated before payment where possible; not charged)
content:
application/json:
schema:
type: object
properties:
error:
type: string
description: Machine-readable error code, e.g. bad_request, not_found, upstream_error, upstream_unreachable, upstream_not_configured
examples:
- bad_request
message:
type: string
description: Human-readable explanation
examples:
- month must be YYYYMM (e.g. 202409)
required:
- error
'402':
description: Payment Required (x402 v2). Returned for every unpaid request. PAYMENT-REQUIRED response header carries base64 JSON PaymentRequired; retry with PAYMENT-SIGNATURE.
'500':
description: Unexpected server error (not charged)
content:
application/json:
schema:
type: object
properties:
error:
type: string
description: Machine-readable error code, e.g. bad_request, not_found, upstream_error, upstream_unreachable, upstream_not_configured
examples:
- bad_request
message:
type: string
description: Human-readable explanation
examples:
- month must be YYYYMM (e.g. 202409)
required:
- error
'503':
description: Endpoint temporarily disabled or upstream not configured on this server; no 402 is issued (not charged). Check /status, retry later
content:
application/json:
schema:
type: object
properties:
error:
type: string
description: Machine-readable error code, e.g. bad_request, not_found, upstream_error, upstream_unreachable, upstream_not_configured
examples:
- bad_request
message:
type: string
description: Human-readable explanation
examples:
- month must be YYYYMM (e.g. 202409)
required:
- error
tags:
- AI
x-x402-network: eip155:8453
x-billing-policy: 'No charge on failure: payment is settled only after the endpoint returns a 2xx response. Any 4xx or 5xx (bad input, no data, upstream error, timeout, disabled endpoint) is not settled, so the signed payment is never collected. Input is validated before the 402 wherever possible. Over MCP, a tool result with isError is not settled either.'
x-error-codes:
- status: 400
codes:
- bad_request
- url_not_allowed
- bad_checksum
- not_a_contract
meaning: Invalid or missing parameters, or a URL/domain that is not allowed.
retry: Fix the input; do not retry as is.
- status: 402
codes: []
meaning: Payment required (x402 v2). The PAYMENT-REQUIRED header holds the exact price.
retry: Pay with an x402 client and retry with PAYMENT-SIGNATURE.
- status: 403
codes:
- not_bok_authored
meaning: Requested table/resource is not served (e.g. statistics not authored by the Bank of Korea).
retry: Do not retry.
- status: 404
codes:
- not_found
- tx_not_found
- tx_not_mined
meaning: No data for the requested period/code yet.
retry: Try an earlier period or another code.
- status: 413
codes:
- too_large
meaning: Document is larger than this server's conversion cap (checked before payment).
retry: Send a smaller document.
- status: 415
codes:
- unsupported_type
- type_mismatch
meaning: Document type not supported or not matching the declared type (before payment).
retry: Do not retry as is.
- status: 502
codes:
- upstream_error
- upstream_bad_response
- upstream_http_error
- rpc_unavailable
- source_http_error
meaning: The upstream source (government API, RPC, target site) returned an error or unreadable data.
retry: Retry after 30-60 s with backoff.
- status: 503
codes:
- endpoint_disabled
- upstream_not_configured
- payment_not_configured
meaning: Endpoint temporarily disabled on this server, or its upstream is not configured. No 402 is issued.
retry: Check /status; retry later.
- status: 504
codes:
- upstream_unreachable
meaning: The upstream source timed out or was unreachable (Korean public data APIs have occasional outages, esp. holidays).
retry: Retry after 1-5 min with backoff.