Use Kimi K3.1 Pro with Claude Code
Claude Code doesn’t require a Claude model: it works with any backend that
speaks the Anthropic Messages API. CheapestInference exposes
Kimi K3.1 Pro through an
Anthropic-compatible endpoint, so you can run
Claude Code on Moonshot’s flagship with unlimited usage during your reserved
time blocks — a flat monthly fee instead of per-token billing. (K3.1 Pro
replaced Kimi K3 in place on November 6, 2026; a
kimi-k3 setup keeps working until December 7, 2026.)
Setup (2 minutes)
Section titled “Setup (2 minutes)”- Subscribe to the Flagship Pool and
copy your subscriber key (
sk-...) from the dashboard. Flagship seats are limited — if the pool is full, join the waitlist. - Export these variables in your shell (or put them in your shell profile):
export ANTHROPIC_BASE_URL="https://api.cheapestinference.com/anthropic"export ANTHROPIC_AUTH_TOKEN="sk-..." # your subscriber keyexport ANTHROPIC_MODEL="kimi-k3.1-pro"export ANTHROPIC_SMALL_FAST_MODEL="kimi-k3.1-pro"- Start
claudeas usual. Every request now runs on Kimi K3.1 Pro.
Why Kimi K3.1 Pro for Claude Code
Section titled “Why Kimi K3.1 Pro for Claude Code”- Flagship-class coding: K3.1 Pro is Moonshot’s strongest model — 55 on the Artificial Analysis Intelligence Index, the closest open-weights experience to the frontier labs’ agents, with a 1M-token context and a natural fit for Claude Code’s plan-then-edit loops.
- Flat price: flagship models elsewhere carry flagship per-token prices, and coding agents multiply them. During your time blocks there is no per-token meter — see Unlimited Subscriptions.
- Thinking blocks: K3.1 Pro’s reasoning surfaces through standard
Anthropic
thinkingcontent blocks, so Claude Code renders it natively. It is off unless Claude Code asks for it — see Reasoning for toggling it.
Per-project configuration
Section titled “Per-project configuration”Pin one project to K3.1 Pro without touching your global shell profile with
.claude/settings.json in the project:
{ "env": { "ANTHROPIC_BASE_URL": "https://api.cheapestinference.com/anthropic", "ANTHROPIC_AUTH_TOKEN": "sk-...", "ANTHROPIC_MODEL": "kimi-k3.1-pro", "ANTHROPIC_SMALL_FAST_MODEL": "kimi-k3.1-pro" }}Can Claude Code use Kimi K3.1 Pro?
Section titled “Can Claude Code use Kimi K3.1 Pro?”Yes. Claude Code talks to any endpoint that implements the Anthropic Messages
API. Point ANTHROPIC_BASE_URL at
https://api.cheapestinference.com/anthropic, set ANTHROPIC_MODEL to
kimi-k3.1-pro, and Claude Code runs on K3.1 Pro — tool calls, streaming and
thinking blocks included. No plugin or proxy on your machine.
Is Kimi K3.1 Pro good for coding agents?
Section titled “Is Kimi K3.1 Pro good for coding agents?”It’s Moonshot’s flagship and the strongest open-weights model for agentic coding — long tool-use chains, big-picture refactors, and instruction following are its strengths. Per-token, flagship models are the expensive tier everywhere (see the LLM API pricing comparison); running K3.1 Pro on a flat-rate block is what makes it viable as a daily driver.
What does it cost to run Claude Code on Kimi K3.1 Pro?
Section titled “What does it cost to run Claude Code on Kimi K3.1 Pro?”A flat monthly subscription to the Flagship Pool — see current prices. Usage during your reserved time blocks is unlimited, so a heavy week costs the same as a light one. Seats are limited.
- Still on
kimi-k3? It is served by K3.1 Pro until December 7, 2026 — switch the two variables tokimi-k3.1-probefore then. - Your Flagship Pool subscription also covers
Qwen3.8 Max with vision input —
switch models by changing
ANTHROPIC_MODEL, no new key needed. - Not ready for flagship? The same setup works with GLM 5.3 (Frontier Pool) or DeepSeek V4.1 Flash (Core Pool).
- General setup, model table and troubleshooting: the main Claude Code guide.