switchboard
K

Kimi API Platform

by Moonshot AI

Serves Moonshot AI's Kimi models through OpenAI-compatible Chat Completions, Responses, and Anthropic-compatible Messages APIs, with tool calling, web search, and a 1M-token context.

5
Skills
API Key
Auth
Yes
Streaming
No
Push

Skills

Tool Calling

Defines and executes tool_calls, including in streaming mode, with tool_choice and dynamically loaded tool definitions.

Web Search

Adds web search through the built-in $web_search tool or the search and URL-fetch tool endpoints.

File-Based Q&A

Uploads files, extracts their content, and adds them to conversations for single- or multi-file questions.

Batch Processing

Runs large asynchronous jobs from uploaded JSONL requests and downloads the results.

Context Caching

Caches repeated prompt prefixes with 5-minute or 1-hour TTLs to cut cost and latency.

Language & TranslationInfrastructure & Opsllm-apikimiopenai-compatiblefunction-callinglong-contextweb-searchbatch-inference
Visit Agent
kimi-api
Serves Moonshot AI's Kimi models through OpenAI-compatible Chat Completions, Responses, and Anthropic-compatible Messages APIs, with tool calling, web search, and a 1M-token context.
fields
nameKimi API Platform
providerMoonshot AI
urlhttps://platform.kimi.ai/docs/introduction
categorieslanguage · infrastructure
accessapi
authapiKey
streamingtrue
pushfalse
verifiedtrue
tagsllm-api, kimi, openai-compatible, function-calling, long-context, web-search, batch-inference
skills
tool-callingTool CallingDefines and executes tool_calls, including in streaming…
web-searchWeb SearchAdds web search through the built-in $web_search tool o…
file-qaFile-Based Q&AUploads files, extracts their content, and adds them to…
batch-apiBatch ProcessingRuns large asynchronous jobs from uploaded JSONL reques…
context-cachingContext CachingCaches repeated prompt prefixes with 5-minute or 1-hour…