Moonshot AI: Kimi K2 Instruct
moonshotai/kimi-k2-instruct
Try Kimi K2 Instruct for free in your browser, and add it to your app for free with Puter.js AI API.
Try it free Add to your app// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';
puter.ai.chat("Explain quantum computing in simple terms", {
model: "moonshotai/kimi-k2-instruct"
}).then(response => {
document.body.innerHTML = response.message.content;
});
<html>
<body>
<script src="https://js.puter.com/v2/"></script>
<script>
puter.ai.chat("Explain quantum computing in simple terms", {
model: "moonshotai/kimi-k2-instruct"
}).then(response => {
document.body.innerHTML = response.message.content;
});
</script>
</body>
</html>
Model Card
Kimi K2 Instruct is the instruction-tuned version of Moonshot AI's Kimi K2, released in July 2025 as a general-purpose chat and agentic model. Moonshot describes it as reflex-grade, meaning it answers directly without extended thinking, which keeps latency low in tool-calling loops.
It scores 65.8% on SWE-bench Verified with agentic tools (single attempt), 70.6% on Tau2-bench retail, and 76.5% on AceBench, close to Claude 4 on tool-use benchmarks at the time of release. Tool calling is built in; you pass a list of available tools and the model decides when and how to invoke them.
This route offers a 128K context window. It suits developers building agents, coding assistants, and multi-step tool pipelines that need autonomous execution without chain-of-thought token overhead.
Context Window 128K
tokens
Max Output 66K
tokens
Input Cost $0.7
per million tokens
Output Cost $2.5
per million tokens
Tool Use Yes
Release Date Jul 11, 2025
Try Kimi K2 Instruct for free
Try Kimi K2 Instruct instantly in your browser.
This playground uses the Puter.js AI API — no API keys or setup required.
More AI Models From Moonshot AI
Find other Moonshot AI models →
Kimi K3.1 Pro
Kimi K3.1 Pro is a multimodal language model from Moonshot AI, announced October 23, 2026 as a post-trained upgrade of Kimi K3. It accepts text, image, and video input, has a 1-million-token context window, and Moonshot plans to release the weights by November 4. Moonshot reports 65.7% on Terminal-Bench 4.0, ahead of GPT-6 Astra and slightly behind Claude Opus 5.5, and 92.7% on BenchCAD, a 3D model coding benchmark. It also reports higher scores than both models on OfficeQA Pro v2, a financial analysis benchmark. Cybersecurity benchmark scores trail both, since Moonshot removed cybersecurity content from post-training data. It suits developers building coding agents, tool-calling pipelines, and financial or document analysis workflows that need long context and multimodal input.
ChatKimi K3.1 Pro HighSpeed
Kimi K3.1 Pro HighSpeed is the faster-serving variant of Kimi K3.1 Pro, the multimodal language model from Moonshot AI announced in October 2026. It accepts text, image, and video input and has a 1-million-token context window. Moonshot says it delivers the same model capabilities as Kimi K3.1 Pro at 2 to 3 times the output speed, for more responsive interactions. It is priced at twice the standard model's rate. Choose it when latency matters more than cost, such as interactive coding agents, tool-calling loops with many sequential steps, and chat or document analysis features where users wait on the response. For batch or background jobs, the standard Kimi K3.1 Pro costs less.
ChatKimi K3
Kimi K3 is Moonshot AI's flagship open-weight model, released July 16, 2026, with full weights following on July 27. At roughly 2.8 trillion parameters in a Mixture-of-Experts architecture, Moonshot positions it as the largest open-source model released to date, built on two new components: Kimi Delta Attention, a hybrid linear attention mechanism, and Attention Residuals, a replacement for standard residual connections. It runs in an always-on thinking mode with a 1-million-token context window and accepts text, image, and video input. Reported results include 93.5% on GPQA Diamond, 91.2% on BrowseComp, 88.3% on Terminal-Bench 2.1, and a first-place finish on Arena.ai's Frontend Code Arena, putting it close to Claude Opus 4.8 and GPT-5.5 on several agentic and coding tasks. These figures come from Moonshot and early testers, not independently confirmed leaderboards. It suits developers building long-horizon coding agents and tool-calling pipelines who want frontier-level performance at open-weight pricing.
Frequently Asked Questions
You can access Kimi K2 Instruct by Moonshot AI through Puter.js AI API. Include the library in your web app or Node.js project and start making calls with just a few lines of JavaScript — no backend and no configuration required. You can also use it with Python or cURL via Puter's OpenAI-compatible API.
Kimi K2 Instruct is free to try with a Puter account. Every account includes a free AI allowance, and you can chat with it in the playground on this page. You can upgrade your account anytime for a larger allowance.
Kimi K2 Instruct is free to integrate using the Puter.js AI API. With the User-Pays Model, you can add AI to your app for $0, since users cover their own AI usage through their Puter account.
| Price per 1M tokens | |
|---|---|
| Input | $0.7 |
| Output | $2.5 |
Kimi K2 Instruct was created by Moonshot AI and released on Jul 11, 2025.
Kimi K2 Instruct supports a context window of 128K tokens. For reference, that is roughly equivalent to 256 pages of text.
Kimi K2 Instruct can generate up to 66K tokens in a single response.
Yes, Kimi K2 Instruct supports tool use (function calling), allowing it to interact with external tools, APIs, and data sources as part of its response flow.
Yes — the Kimi K2 Instruct API works with any JavaScript framework, Node.js, or plain HTML through Puter.js. Just include the library and start building. See the documentation for more details.
Add Kimi K2 Instruct to your app for free
Developers can integrate Kimi K2 Instruct for free using the Puter.js AI API.
With the User-Pays Model, each user covers their own AI usage instead of the developer.