We pin the Kimi K2 0905 API in production — a fixed snapshot means our agents do not drift when an upstream model updates.
The Kimi K2 0905 API is Moonshot AI's pinned September 2025 snapshot of the open-weight Kimi K2 — a fixed, reproducible checkpoint with the 0905 update's stronger coding and agentic ability, plus native tool calling, structured outputs, and a long context window. The Kimi K2 0905 API is OpenAI-compatible: point any existing SDK at RouterBase and set the model to kimi-k2-0905. Input at $0.51 / 1M tokens, output at $2.125 / 1M tokens — 15% below the official rate, all through one REST endpoint.
Input
Kimi K2 0905
Hi! I'm a helpful AI assistant. What can I do for you?
Kimi K2 0905 API: Chat
Use the Kimi K2 0905 API to pin Moonshot AI's September 2025 Kimi K2 snapshot — a fixed, reproducible open-weight checkpoint carrying the 0905 update's sharper agentic coding and tool use.
The Kimi K2 0905 API is Moonshot AI's pinned September 2025 Kimi K2 snapshot — a fixed, reproducible open-weight checkpoint with the 0905 update's stronger coding and agents, plus native tool calling, structured outputs, and a long context window. Routed through RouterBase, the Kimi K2 0905 API is fully OpenAI-compatible: point any existing SDK at RouterBase's base URL, set the model to kimi-k2-0905, and your first Kimi K2 0905 API call goes through immediately.
Pricing for the Kimi K2 0905 API is $0.51 / 1M input and $2.125 / 1M output — 15% below the official rate, via one RouterBase key (no Moonshot AI account).
Six reasons teams ship on the Kimi K2 0905 API
From a pinned, reproducible checkpoint to a 15%-off price — why teams standardize on the Kimi K2 0905 API.
A pinned, reproducible snapshot
Kimi K2 0905 is a date-stamped checkpoint — pin it and behavior stays put, so the evals, prompts, and agents you tuned today still work the same next quarter.
The 0905 coding update
The 0905 release sharpened Kimi K2 on coding and agentic ability — stronger multi-file edits, tool use, and front-end work than the earlier K2 checkpoints.
Native tool calling
The Kimi K2 0905 API does function / tool calling in the OpenAI format — Kimi K2 was built around tool use, so agents drop in without custom glue.
Long context window
The Kimi K2 0905 API takes a long context — feed it whole repositories, long documents, or a full agent trajectory in one call.
OpenAI-compatible, one key
The Kimi K2 0905 API uses the OpenAI chat-completions format, so one RouterBase key swaps Kimi K2 0905 in beside GPT-5.5, Claude, Gemini, and 200+ other models.
15% off the list price
Kimi K2 0905 runs 15% under Moonshot AI's published rate — $0.51 / 1M input, $2.125 / 1M output. Open-weight pricing on a pinned version, billed per token.
Quickstart
Get started with the Kimi K2 0905 API in 3 steps
From sign-up to your first response in under 5 minutes.
Create a RouterBase API key
Sign up and create an API key — it reaches Kimi K2 0905 and every other model in the RouterBase catalog from one account.
Send your first message
Point any OpenAI-compatible SDK at routerbase.com/v1 and set the model to kimi-k2-0905 to pin this exact snapshot. Streaming, tool calls, and structured outputs all follow the standard chat-completions schema.
Inspect usage
Every Kimi K2 0905 API response returns a token breakdown — input and output — so cost stays visible on every call.
Pay only for what you use
RouterBase passes through partner-tier pricing. Compared against the model's official published API rate.
What teams build with the Kimi K2 0905 API
Real production loads running on the RouterBase model catalog across 200+ models.
We standardized on Kimi K2 0905 through RouterBase so our evals stay reproducible — same checkpoint, same numbers, every release.
The 0905 update noticeably improved tool-call reliability for our agents — fewer retries than the earlier K2 we ran.
We A/B the Kimi K2 0905 API against GPT and Claude behind one RouterBase key — on coding tasks it holds up for far less.
Pinning to kimi-k2-0905 was one base-URL and model-id change — a stable version we control, without leaving our OpenAI SDK.
Multi-file refactors got sharper on the 0905 snapshot — we pushed our coding agents to the Kimi K2 0905 API and kept the wins frozen.
As a solo dev I pin Kimi K2 0905 so my app behaves the same months later — open weights, fixed checkpoint, cheap to run.
Reproducibility is why we run the Kimi K2 0905 API — a dated snapshot keeps our regression suite meaningful.
Per-token billing plus a pinned Kimi K2 0905 means predictable cost and predictable behavior — both matter for our audits.
Frequently Asked Questions
Common questions about the Kimi K2 0905 API.
The Kimi K2 0905 API is RouterBase's pass-through to Moonshot AI's Kimi K2 0905 — a pinned September 2025 snapshot of the open-weight Kimi K2 carrying the 0905 update's stronger coding and agents, served over an OpenAI-compatible REST interface with native tool calling and structured outputs.