Deterministic tools for AI agents.
Deterministic tools for AI agents.
More reliable than AI inference. A lightweight utility API suite for developers and AI agents.
{
"mcpServers": {
"thousand-api": {
"url": "https://mcp.thousand-api.com/mcp",
"headers": { "x-api-key": "your-api-key" }
}
}
}Save LLM tokens and ensure accurate results
When AI agents use tools, they get accurate results with fewer tokens than having the LLM process everything itself.
Two approaches to reducing tokens
① Use tools to reduce output tokens
e.g. 95% reduction with HTML→text extraction
② Turn off unused tools to reduce tool-definition tokens
- Configure per category or per tool from the dashboard
- All 55 tools → 20 tools cuts tool-definition tokens by about 64%
Common filtering presets:
- Developer set (6 tools) → Security + Data processing
- Document set (5 tools) → Text + Format conversion
- Research set (5 tools) → Network
94%
Token reduction
HTML to text
64%
Token reduction
Postal code lookup
67%
Token reduction
Compound interest
60%
Token reduction
HMAC verification
Without tools
Total: 9,300 TOKEN
With tools
Total: 540 TOKEN
↑ 94% saved
What AI agents struggle with—done reliably
Close LLM weak spots we’ve verified in practice—with tools that don’t guess.
Live data
LLMs only know what they were trained on
- DNS lookup: Misstated GitHub’s MX as “Gmail” (correct: Microsoft 365)
- FX rates: Guessed a range like “~161–162 yen” (Thousand API returns exact values)
- URL redirects: Invented a fictional destination for shortened URLs
Reliable generation
LLMs invent plausible-looking values
- SHA-256 / HMAC: Produced fake hashes that only look correct
- ISBN validation: Marked every valid ISBN-13 as invalid
- UUID generation: No cryptographic randomness guarantee
Boundary-sensitive math
LLMs only approximate the hard edges
- Business days: Off-by-one from how the start date is counted
- JWT expiry: Unix time conversion off by 6+ hours
- Map tile coordinates: ~10 tiles off on the Y axis
AI makes mistakes. We tested it.
We asked the same questions to a general AI chat and Thousand API (MCP-connected).
🔐 SHA-256 hash generation
General AI chat
aa69305e7a9… (a value that was not actually correct)
Thousand API
Returns the correct SHA-256 hash instantly
🌐 DNS lookup (github.com MX)
General AI chat
aspmx.l.google.com (stale training data)
Thousand API
outlook.com (the live MX record)
🔗 Short URL resolution
General AI chat
Confidently answered with a URL that does not exist
Thousand API
Follows HTTP redirects and returns the final URL
📅 Business days (10 days after June 27)
General AI chat
July 28 (off by one from inclusive start-date counting)
Thousand API
July 29 (exact, per the specification)
July 2026 / Compared with ChatGPT Free
Use cases
Hand agents MCP tools for work that breaks when left to inference
AI assistant
Scheduling for AI assistants
Meeting times and reminders drift when agents infer business days, time zones, or cron schedules.
Related tools
- Holiday & Business Day
- Cron Parser
AI marketing
Prep work for AI writers and marketers
Outlines and competitor research suffer when meta tags and post timing are guessed.
Related tools
- Page Meta / OGP
- QR Code Generator
- Holiday & Business Day
Cursor / Claude dev
Precise utilities for AI engineers
JSON, cron, and URLs in generated code introduce subtle spec mistakes.
Related tools
- JSON Validator
- Cron Parser
- URL Redirect Resolver
Chatbot
External sensors for chatbots
Bots lack live locale, link targets, and final URLs for short links.
Related tools
- IP Geolocation
- Page Meta / OGP
- QR Code Generator
- URL Redirect Resolver
On “10 business days later,” typical LLMs often miscount the start date. Thousand API calculates exactly to the specification.
Code generation, text processing, and Japanese conversion with reliable, token-efficient APIs
Format conversions for AI-generated data with zero tokens
Extract values from large JSON, validate schemas, compute statistics
Typical LLMs pretend to know DNS records and URL contents. Thousand API actually fetches and verifies them.
SHA-256 and HMAC invite LLMs to invent plausible fake values. Thousand API performs real cryptographic computation.
Distance, coordinates, color tools, and version comparison with deterministic results
Tool names, endpoints, parameters, and response examples are in the
MCP tool index →Works as an MCP server
Connect directly to Claude or Cursor. Copy the config and your AI agent can call Thousand API tools right away.
Supported clients
- Claude
- Cursor
- Other MCP-compatible tools
{
"mcpServers": {
"thousand-api": {
"url": "https://mcp.thousand-api.com/mcp",
"headers": { "x-api-key": "your-api-key" }
}
}
}Pricing
50+ MCP tools free for up to 1,000 requests/month. Upgrade as you scale.
Free
Free
Trial and light personal use
- MCP server connection
- REST API access
- API keys1
- Monthly requests1,000
- Email support
- Priority support
Starter
¥480/mo
Everyday personal use
- MCP server connection
- REST API access
- API keys5
- Monthly requests10,000
- Email support
- Priority support
Pro
¥1,980/mo
Power users and small teams
- MCP server connection
- REST API access
- API keys20
- Monthly requests100,000
- Email support
- Priority support