Kimi K2 0905

Provider: Moonshot AI (Kimi) CN

Input price
$0.60
/ 1M tokens
Output price
$2.50
/ 1M tokens
Context
262K tokens
Max output 98K tokens

Overview

Moonshot AI's Kimi K2 0905 is a mid-to-high tier model designed for workloads requiring extensive context processing. It provides a massive context window and robust capabilities for structured output. This model targets applications needing to process and generate very long sequences of text or highly structured data.

Strengths

  • Exceptional 262,144 token context window for both input and output.
  • Reliable function-calling capabilities for complex application integration.
  • Dedicated JSON mode for structured and predictable responses.
  • Competitive input pricing relative to several peers in its category.
  • Supports processing and generating extremely long text sequences.

Best use cases

  • Analyzing large legal documents, research papers, or extensive codebases.
  • Developing complex agents that require structured interaction with tools.
  • Summarizing extensive logs, detailed reports, or long conversation histories.
  • Generating structured data or code snippets based on comprehensive instructions.
  • Orchestrating multi-step workflows with external tools via function calls.

Pricing position

Kimi K2 0905 is priced at $0.6/1M tokens for input and $2.5/1M tokens for output, matching its sibling Kimi K2 Thinking. Its input price is competitive with several Zhipu GLM models, but its output price is notably higher than GLM 5 ($1.92/1M), GLM 4.5 ($2.2/1M), and especially Gemma 2 27B ($0.65/1M). It offers a more cost-effective output than Kimi K2.6 ($3.49/1M) and Qwen3 Coder Plus ($3.25/1M).

Caveats

There is no free tier available for Kimi K2 0905, which might be a consideration for initial testing or low-volume use cases. While powerful, its output price is higher than some similarly-priced input peers, suggesting that for tasks not requiring its massive context or structured output features, cheaper models like Kimi K2.5 or GLM 5 might be more economical. This model does not explicitly list vision capabilities.

Notes drafted by an LLM from the figures on this page (2026-05-21) and rewritten when the price changes.

Key facts

Price basis
Median third-party host price (no first-party API)
Max output
98K tokens
Knowledge cutoff
2024-12-31
Listed on OpenRouter
2025-09-04
Tracked since
2026-05-09
Last updated
2026-11-10

Features

Function callingJSON output

Monthly cost estimate

Scale Input (M) Output (M) Monthly
0.1M (hobby) 0.1 0.05 $0.18
1M (small app) 1 0.5 $1.85
10M (mid scale) 10 5 $18.50
100M (large) 100 50 $185.00

※ 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-k2-0905)