Kimi K2 Thinking: What It Does and Costs
moonshotai/kimi-k2-thinkingKimi K2 Thinking is an open-weight text model for reasoning and tool use.
262K tokens of context — about 196,608 words, on a scale that ends at 2M.
What it can do
- Reasoning
- Tool use
- Open weights
- Takes text
- Gives text
Released 6 Nov 2025 · checked 3 Oct 2026.
About it
Made by MoonshotAI, Kimi K2 Thinking is a text-only reasoning model with open weights and tool support. It can take up to 262,144 tokens of text at once and produce up to 235,929 tokens in one answer. Its listed prices are $0.6 per million input tokens and $2.5 per million output tokens. It supports tools.
Review
In a line
Kimi K2 Thinking is a text-only model from MoonshotAI. Its weights are open, and it supports reasoning and tools.
What it's good at
This model is suited to tasks that benefit from reasoning over text and using tools. It can process text input and produce text output.
Price in plain terms
The price is $0.6 per million input tokens and $2.5 per million output tokens. A token is a small piece of text. Reading a 500-word page costs about 0.039 cents; a chat turn with 1,000 input tokens and 500 output tokens costs about 0.18 cents. A thousand such turns cost about $1.85.
Context and limits
It can take up to 262,144 tokens in one request and write up to 235,929 tokens in one answer. It handles text only; the listed inputs and outputs are text.
Ways to use it
It supports tools. Its weights are open, so it is also listed among open-weight models.
Verdict
Consider Kimi K2 Thinking if you want a text model for reasoning and tool use, need a large text window, or want open weights. Its token prices are listed for the sample tasks, and actual costs depend on how much text goes in and comes out. Skip it if you need image input or output, since it handles text only.
Verdict
Kimi K2 Thinking may suit people comparing text models for reasoning tasks, especially if they value open weights, tool support, and room for a large amount of text in one request. Its prices are listed separately for text going in and answers coming out, so longer responses add cost. The sample estimates are tied to specific token amounts. Choose another model if your work depends on images rather than text.
- Open weights are available.
- Supports reasoning and tools.
- Accepts up to 262,144 tokens at once.
- Listed prices are $0.6 in and $2.5 out per million tokens.
- Text only; it does not handle image input or output.
- Output is billed at a higher per-million-token rate than input.
- Pick it ifText tasks that benefit from reasoning and tool use; People who want an open-weight model
- Skip it ifYou need a model that accepts or produces images; You need a smaller context window or shorter answers
Pricing
Costs $0.6 per million input tokens and $2.5 per million output tokens. The supplied examples estimate about 0.039 cents to read a 500-word page, about 0.18 cents for a sample chat turn, and about $1.85 for 1,000 such turns.
Questions
Who makes Kimi K2 Thinking?
MoonshotAI makes the model.
Is Kimi K2 Thinking open weight?
Yes. Its weights are open.
Can it work with images?
No. It takes text input and produces text output.
How much text can it handle at once?
Its context window is 262,144 tokens.
How long can its answer be?
It can produce up to 235,929 tokens in one answer.
What does it cost?
The listed rates are $0.6 per million input tokens and $2.5 per million output tokens.
Does it support tools?
Yes. It supports tools.