moonshotai

MoonshotAI: Kimi K2 0905

The MoonshotAI Kimi K2 0905 is a text-based AI model that can handle context lengths of up to 262144 tokens and supports tool integration. It does not possess reasoning capabilities or output structured data. This model's price point may be worth considering for organizations requiring text input tools, as it offers a relatively affordable input token rate of $0.6 per million tokens. However, its blended benchmark score of 23.2 across one independent benchmark indicates that it lags behind other models in this area; further evaluation is necessary to determine if the MoonshotAI Kimi K2 0905's strengths justify its cost.

Quality Score
93/100
price + capability + benchmarks
Input Price
$0.60
per 1M tokens · checked 2026-09-10
Output Price
$2.50
per 1M tokens · checked 2026-09-10
Context Window
262,144
tokens

Benchmark results

Independent, published benchmarks. Blended score 23.2 across 1 benchmark, last refreshed 2026-09-10. How scoring works →

BenchmarkMeasuresScore
AI Index broad capability composite 25.2
Model ID
moonshotai/kimi-k2-0905
Vendor
moonshotai
Released
September 2025
Tokenizer
Other
Input Modalities
text
Output Modalities
text
Max Output
100,352 tokens
Tool Calling
✓ supported
Structured Output
✓ supported
Reasoning Mode
not supported
Vision
text only
Audio
no
Moderated
no

What it costs in practice

Computed from the current $0.60/M input and $2.50/M output rates. Run your own numbers →

JobTokensCost
Summarize a 50-page report 30k in / 1.5k out $0.02
Classify 1,000 customer emails 500k in / 50k out $0.42
A month of a busy support chatbot 5M in / 2M out $8.00

Similar models

Quick answers

How much does MoonshotAI: Kimi K2 0905 cost?
$0.60 per million input tokens and $2.50 per million output tokens.
What is MoonshotAI: Kimi K2 0905's context window?
262,144 tokens, roughly 393 pages of text in a single request.
Does MoonshotAI: Kimi K2 0905 support tool calling?
It supports tool calling, structured output.
Can MoonshotAI: Kimi K2 0905 process images?
No, it is text-only on the input side.