Skip to content
CheapestInference

Kimi K3 API — pricing, access & subscription

Kimi K3 (also written “Kimi 3”) is Moonshot’s flagship agentic and coding model — the first open-weight model to outscore the Claude Opus tier on the Artificial Analysis Intelligence Index. CheapestInference serves it in the Flagship Pool on unlimited, flat-rate time-block subscriptions from $149/mo — no per-token billing, no token caps during your reserved hours. Seats are very limited; live availability is on /pools.

Quick facts

Section titled “Quick facts”
ModelKimi K3
Model IDkimi-k3
Context window256K tokens
PoolFlagship Pool (sole model)
Per-token price elsewhere$3.00 in / $15.00 out per 1M — reference only
Endpoints/v1/chat/completions (OpenAI), /anthropic/v1/messages (Anthropic)
PricingFrom $149/mo — reserve an 8-hour daily time block, up to full 24/7

Call Kimi K3

Section titled “Call Kimi K3”
from openai import OpenAI


client = OpenAI(
    base_url="https://api.cheapestinference.com/v1",
    api_key="sk-..."  # your subscriber key
)


response = client.chat.completions.create(
    model="kimi-k3",
    messages=[{"role": "user", "content": "Refactor this function..."}],
)
Terminal window
curl https://api.cheapestinference.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "kimi-k3", "messages": [{"role": "user", "content": "Hello"}]}'

Reasoning is off by default — you get fast, direct answers. Enable it per request with "thinking": { "type": "enabled" }, or with reasoning_effort — K3 natively supports low, high, and max, its deepest tier. With reasoning on, set a generous max_tokens so answers are not truncated mid-reasoning.

Why unlimited Kimi K3

Section titled “Why unlimited Kimi K3”

Per token, K3 is one of the most expensive open-weight models on the market ($3/$15 per 1M elsewhere) — exactly the profile where a flat fee pays off. A Flagship Pool subscription gives you K3 through a standard OpenAI-compatible endpoint with no token caps: heavy agentic coding runs, long sessions, full workdays, all at one fixed monthly price. It works in Claude Code, Cline, Roo Code, Continue, and any client that accepts a custom base URL.

Because seats are very limited, the pool can sell out — when it does, the pools page shows live availability per block.

Common questions

Section titled “Common questions”

Is there an unlimited Kimi K3 API? Yes. The CheapestInference Flagship Pool serves Kimi K3 with uncapped usage for one flat monthly fee — from $149/month for a daily 8-hour block, up to full 24/7 with all three blocks. Seats are very limited.

What is the cheapest way to use Kimi K3? For heavy use, an unlimited time-block subscription: from $149/month you reserve a daily 8-hour window and use K3 with no token caps. At K3’s per-token list price ($3 in / $15 out per 1M), a serious coding-agent workload passes the subscription price within the month. For light or occasional use, per-token access via Moonshot’s API is the cheaper route.

Is Kimi K3 OpenAI-compatible? Yes. Use model id kimi-k3 against https://api.cheapestinference.com/v1 with the OpenAI SDK, or the /anthropic endpoint with the Anthropic SDK — which also makes it work in Claude Code.

Is there a Kimi K3 coding plan? Yes — a Flagship Pool subscription works as a K3 coding plan: unlimited Kimi K3 through Claude Code, Cline, Roo Code or any OpenAI-compatible coding agent, with no weekly request quotas to manage.

How is this different from the Frontier Pool? The Frontier Pool bundles Kimi K2.7, GLM 5.2 and MiniMax M3 at a lower price. The Flagship Pool serves K3 only — Moonshot’s current flagship — with very limited seats.

Section titled “Related”