Moonshot AI · Model
Kimi K3 pricing, context window and specs
Kimi K3 costs $0.53 per million input tokens and $12 per million output tokens, and reads up to 1,048,576 tokens per request. By blended price it is the most expensive in Moonshot AI's current line-up.
- Input / 1M
- $0.53
- Output / 1M
- $12
- Cached / 1M
- $0.30
- Context
- 1,048,576
- Max output
- Not published
- Released
- 2026-07-16
Estimate your cost with Kimi K3Compare all Moonshot AI modelsCount tokens for Kimi K3
Pricing
Kimi K3 API pricing
| Price per 1M tokens | Input | Output |
|---|---|---|
| Standard | $0.53 | $12 |
| Cached input (read) | $0.30 | – |
| Batch API | $2.28 | $11.40 |
Output costs 22.6× the input price, so long replies drive the bill more than long prompts. If your requests share a long fixed prefix, such as a system prompt or tool definitions, caching cuts that part to 57% of the normal input price.
Examples
What Kimi K3 costs in practice
| Workload | Tokens in / out | Requests/day | Per request | Per month |
|---|---|---|---|---|
| Support chatbot50% cached | 1,500 / 400 | 1,000 | $0.00542 | $164.93 |
| RAG app20% cached | 6,000 / 500 | 500 | $0.0089 | $135.42 |
| Coding agent80% cached | 40,000 / 2,000 | 200 | $0.0378 | $230.19 |
| Document summariserbatch | 8,000 / 600 | 300 | $0.0251 | $228.86 |
Change any number in the cost calculator.
Context
Context window
Kimi K3 accepts up to 1,048,576 tokens per request, roughly 786,432 English words or 1,573 pages. That budget is shared between your prompt and the reply, and no separate cap on the reply is published.
Filling the whole window with one prompt costs about $0.56 at list price, so for repeated long-document work, retrieval (sending only the relevant parts) or prompt caching is usually much cheaper.
Features
Capabilities
- Image input
- Yes
- Tool / function calling
- Yes
- Reasoning mode
- Yes
- Prompt caching
- Yes
- Structured output
- Yes
- Open weights
- Yes
- Accepts
- text, image, video
- Knowledge cutoff
- Not published
As declared by the provider’s API. It shows a feature exists, not how well it works.
Moonshot AI
Where Kimi K3 sits in Moonshot AI’s line-up
Ranked by blended price (three parts input to one part output, $3.3975 per 1M for Kimi K3), it is the most expensive of Moonshot AI’s 7 priced models. See every Moonshot AI model and price
| Model | Input / 1M | Output / 1M | Context | Released |
|---|---|---|---|---|
| Kimi K3 (this page) | $0.53 | $12 | 1,048,576 | 2026-07-16 |
| Kimi K2.7 Code | $0.6712 | $3.35 | 262,144 | 2026-06-12 |
| Kimi K2.6 | $0.465 | $2.45 | 262,144 | 2026-04-20 |
| Kimi K2.5 | $0.45 | $2.25 | 262,144 | 2026-01-27 |
| Kimi K2 Thinking | $0.60 | $2.50 | 262,144 | 2025-11-06 |
| Kimi K2 0905 | $0.60 | $2.50 | 262,144 | 2025-09-04 |
| Kimi K2 0711 | $0.57 | $2.30 | 131,072 | 2025-07-11 |
Save
Cheaper alternatives
- Qwen3.6 35B A3BQwen89% cheaper
- GLM 4.6VZ.ai87% cheaper
- GPT-5.6 LunaOpenAI87% cheaper
- DeepSeek V4.1 FlashDeepSeek85% cheaper
- Muse Glimmer 30BMeta85% cheaper
Current models from major providers, one per provider, that keep image input, tool calling, a reasoning mode, a usable context window and output limit. Compared by blended price.
Compare
Similarly priced models
- Gemini 3.5 FlashGoogle$1.50 / $9
- GPT-5.1OpenAI$1.25 / $10
- Mistral Medium 3.5Mistral$1.50 / $7.50
- Qwen3.8 2.4T A95BQwen$2 / $6
- Grok 4.5xAI$2 / $6
Current models from other major providers, closest blended price.
FAQ
Kimi K3 questions
How much does Kimi K3 cost?
Kimi K3 costs $0.53 per million input tokens and $12 per million output tokens, with cached input at $0.30. A typical chatbot reply (1,500 tokens in, 400 out) costs about $0.00559. Use the LLM cost calculator for your own workload.
What is Kimi K3's context window?
1,048,576 tokens, about 786,432 English words or 1,573 printed pages, shared between your prompt and the reply. No separate limit on the reply is published, so the window is the limit.
Does Kimi K3 support prompt caching and batch requests?
Yes, cached input is billed at $0.30 per million tokens (57% of the normal input price). A batch API is available at $2.28 input and $11.40 output per million tokens.
What are cheaper alternatives to Kimi K3?
With the same essentials (image input and tool calling), the cheapest options are Qwen3.6 35B A3B (89% cheaper), GLM 4.6V (87% cheaper), GPT-5.6 Luna (87% cheaper). Cheaper doesn't mean equivalent, so test them on your own prompts.
Is Kimi K3 open source?
Yes, Kimi K3 has open weights, so you can run it yourself or choose between several hosting providers. The prices here are a typical hosted price.