Moonshot

Kimi K3

moonshotai/kimi-k3Hugging Face ↗

Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at...

toolsstructuredreasoningimage in
Data source

Only tracked on Vercel AI Gateway.

Kimi K3 usage over time

Vercel AI Gateway share of tokens, by day.

View
Tidelines — LLM usage and market share

Kimi K3 holds 6.86% today, up from 0.00% at the start of the visible window.

Data: Vercel AI Gateway as of 2026-09-04 · Source: tidelines.ai/models/moonshotai-kimi-k3

Rank history

Tidelines — LLM usage and market share
Data: Vercel AI Gateway as of 2026-09-04 · Source: tidelines.ai/models/moonshotai-kimi-k3

Adoption curve vs the pantheon

Kimi K3's share plotted against days since first tracked, overlaid on other current leaders on the same axis. The question every new-model page asks: is this one going to make it?

Tidelines — LLM usage and market share
Data: Vercel AI Gateway as of 2026-09-04 · Source: tidelines.ai/models/moonshotai-kimi-k3

Citable stat

As of 2026-09-04, Kimi K3 holds 6.9% of Vercel AI Gateway token share, up 3.56pp over 30 days. Source: Tidelines (Vercel AI Gateway).

Cost per session

Median dollars spent on one Kimi K3 session, by coding agent and session length.

Tidelines — LLM usage and market share

Token rate sets the price; session length and reasoning set the bill.

Agent1 turn2–9 turns10–49 turns50+ turns
Claude Code$0.073$0.164$1.06$6.34
Kilo Code$0.052$0.233$1.05$5.77
Hermes Agent$0.055$0.160$0.829$4.69
Codex$0.029

Compare every model's session cost →

Data: OpenRouter session-cost dataset as of 2026-08-30 · Source: tidelines.ai/models/moonshotai-kimi-k3

Where to run it

Kimi K3 pricing by provider

Tidelines — LLM usage and market share

Per-provider pricing, context, and uptime via OpenRouter. Sorted by output price.

Kimi K3 is served by 18 tracked endpoints on OpenRouter. The cheapest output price is $12.75 per million tokens from Makora, and the most expensive is $22.50 from Morph · 1049k. The largest context window offered is 1049k tokens. Prices are list rates and refresh daily.

ProviderInput $/MOutput $/MContextUptime
Makora$2.55$12.751049k100.00%
Wafer$3.00$12.751049k
Sail Research$2.60$13.00975k99.96%
Morph · 1049k$2.50$14.001049k99.01%
DeepInfra$2.85$14.251049k97.87%
DigitalOcean$2.85$14.251049k99.98%
Phala$3.00$15.001049k82.03%
Chutes$3.00$15.001049k98.51%
Parasail$3.00$15.001049k92.45%
Modal$3.00$15.001049k99.80%
Together$3.00$15.001049k98.60%
Fireworks · 1049k$3.00$15.001049k99.77%
BaseTen$3.00$15.001049k99.35%
Moonshot AI$3.00$15.001049k99.98%
Fireworks · 1049k$3.30$16.501049k100.00%
Alibaba$3.45$17.251049k96.48%
Fireworks · 1049k$4.50$22.501049k98.06%
Morph · 1049k$6.00$22.501049k97.69%