Kimi K2.5

Provider: Moonshot AI (Kimi) CN

Input price
$0.57
/ 1M tokens
Output price
$2.85
/ 1M tokens
Cache price
$0.095
/ 1M tokens (read)
Context
262K tokens
Max output 236K tokens

Overview

Moonshot AI's Kimi K2.5 is a multimodal flagship model with a 262k context window and native thinking capabilities.

Strengths

  • Supports vision, function calling, and JSON mode
  • Outputs up to 65536 tokens per request
  • Offers cached input pricing at $0.095 per million tokens

Best use cases

  • Complex visual coding tasks and software development
  • Multi-step agent workflows requiring reasoning steps
  • Processing large multimodal documents within 262k context

Pricing position

Input costs $0.57/M and output costs $2.85/M, matching the Kimi K2 0711 baseline while undercutting the newer K2.6 at $0.95/$4 per million.

Caveats

Retirement is scheduled for November 18, 2026. No free tier is available.

Notes drafted by an LLM from the figures on this page (2026-11-10) and rewritten when the price changes.

Key facts

Price basis
Median third-party host price (no first-party API)
Max output
236K tokens
Listed on OpenRouter
2026-01-27
Tracked since
2026-05-09
Last updated
2026-11-12

Features

Image inputFunction callingJSON outputReasoning

Monthly cost estimate

Scale Input (M) Output (M) Monthly
0.1M (hobby) 0.1 0.05 $0.20
1M (small app) 1 0.5 $2.00
10M (mid scale) 10 5 $19.95
100M (large) 100 50 $199.50

※ Cache and batch discounts not factored in. Detailed calculator

Official resources

Other models from Moonshot AI (Kimi)

Similar-priced models

Source: OpenRouter API (moonshotai/kimi-k2.5)