Kimi logo

Kimi

chatbot

by moonshot-ai

Powerful AI assistant with Agent Swarm, 256K context, visual coding, and plans from free to $199/mo.

Key features

Conversations
Agent Mode
Agent Swarm
Four Operating Modes
256K Context Window
Vision and Multimodal Understanding
Pricing

Free tier available, Allegretto at $39/mo

Best For

Anyone who wants a free, capable AI chatbot with 256K context, web search, file uploads, and deep research built in

Verdict

Frontier level performance at a fraction of the cost of GPT-5.2, Claude Opus 4.6, and Gemini Pro 3.1

What it does

Conversations

Ask questions, brainstorm ideas, get explanations on any topic in English or Chinese with strong multilingual support.

Agent Mode

Autonomous task execution where Kimi plans, searches the web, writes code, and iterates to complete complex multi-step requests.

Agent Swarm

Orchestrate up to 100 parallel sub-agents that coordinate across 1,500 steps to solve large scale research, analysis, and coding tasks.

Learn more

Four Operating Modes

Switch between Instant (fast responses), Thinking (deeper reasoning), Agent (autonomous tasks), and Agent Swarm (parallel multi-agent orchestration).

Learn more

256K Context Window

Process very long documents, entire codebases, or extended conversation histories in a single session.

Vision and Multimodal Understanding

Analyze images and video frames with state of the art multimodal capabilities. K2.5 leads on 9 out of 17 image and video benchmarks.

Learn more

Kimi Code CLI

Open source command line coding agent similar to Claude Code. Reads, writes, and modifies code across your project with MCP support.

Learn more

Web Search and Deep Research

Autonomous web browsing and source synthesis that produces detailed reports with citations for comprehensive investigations.

Kimi Claw

Your always-on AI assistant that lives in your browser. Runs 24/7 with persistent memory, 5,000+ ready-made skills, 40 GB cloud storage, and scheduled automations. Built on the open source OpenClaw framework.

Learn more

Pricing

Free

Free

Basic conversations with standard usage limits. Good for trying Kimi and light everyday use.

  • Basic conversations
  • Standard usage limits
  • 256K context window
  • Web and mobile access

Moderato

$19/month

Extended agent quota, multi-tasking, 4x speed agents, K2.5 with higher quota, Slides visual mode, and Kimi Code access.

  • Extended agent quota
  • Agent multi-tasking
  • Agents with 4x speed and priority access
  • K2.5 with higher speed and extended quota
  • Slides visual mode with Nano Banana
  • 3x Boost Kimi Code available
Best Value

Allegretto

$39/month

2x agent quota, 2x K2.5 usage, unlimited Slides, 5x Kimi Code quota, AI Assistant deployment, and Agent Swarm access.

  • Everything in Moderato
  • 2x agent quota
  • 2x usage quota for K2.5 model
  • Unlimited Slides visual mode
  • 5x Kimi Code quota
  • AI Assistant: Deploy Kimi Claw in one click
  • Agent Swarm (Research Preview)

Allegro

$99/month

5x agent quota, 2x multi-tasking, priority access during peak hours, 5x K2.5 usage, 15x Kimi Code quota, and Agent Swarm.

  • Everything in Moderato
  • 5x agent quota
  • 2x agent multi-tasking
  • Priority access during peak hours
  • 5x usage quota for K2.5 model
  • Unlimited Slides visual mode
  • 15x Kimi Code quota
  • AI Assistant and Agent Swarm

Vivace

$199/month

10x agent quota, 10x K2.5 usage, 30x Kimi Code quota, and all premium features. For power users who rely on Kimi daily.

  • Everything in Moderato
  • 10x agent quota
  • 2x agent multi-tasking
  • Priority access during peak hours
  • 10x usage quota for K2.5 model
  • Unlimited Slides visual mode
  • 30x Kimi Code quota
  • AI Assistant and Agent Swarm

Pros & Cons

Pros

  • Frontier level performance at a fraction of the cost of GPT-5.2, Claude Opus 4.6, and Gemini Pro 3.1
  • Open source weights under Modified MIT License; you can self-host, fine-tune, and inspect the model
  • Agent Swarm is a genuinely novel capability: up to 100 parallel agents coordinating without predefined roles
  • Best in class multimodal vision with highest scores on 9 out of 17 image and video benchmarks versus top competitors
  • 256K token context window handles very long documents and entire codebases in a single session
  • Available globally with English interface, USD API pricing, and distribution through US based providers (Fireworks, OpenRouter)
  • Kimi Code provides a free, open source CLI coding agent comparable to Claude Code with MCP support
  • Input caching at $0.10 per million tokens dramatically reduces costs for repeated or iterative tasks

Cons

  • Newer and less established than ChatGPT, Claude, or Gemini with a smaller community and fewer integrations
  • Consumer app ecosystem is less mature; fewer features like memory, projects, or connectors compared to Claude or ChatGPT
  • WeirdML benchmark score of 46% suggests weaker performance on unusual or edge case reasoning compared to competitors
  • Agent Swarm is still in beta; expect rough edges and inconsistent results on some tasks
  • Documentation and support resources are less comprehensive than those from OpenAI or Anthropic
  • The company is based in Beijing, which may raise data residency or regulatory concerns for some organizations (though US based API providers like Fireworks mitigate this)
  • Paid consumer plan details are less transparent than competitors; pricing may vary by region
  • Consumer app writing quality does not yet match Claude for nuanced, natural sounding prose

How to get started

1

Create your account

Go to kimi.com and create a free account. The interface is available in English. You can also download the iOS or Android app for mobile access. The free tier (Adagio) provides unlimited basic conversations with limited Deep Research and Agent tasks.

2

Send your first message and explore operating modes

Type a question, upload a document, or describe a task. Switch between Instant (fast responses), Thinking (deeper reasoning), Agent (autonomous task completion), and Agent Swarm (parallel multi-agent orchestration) using the mode selector. Try Agent mode for research tasks or Thinking mode for complex analysis.

3

Try key features

Activate Agent Swarm (beta) for large scale tasks that benefit from parallel investigation. Use Deep Research for autonomous web research with source citations. For developers, sign up at platform.moonshot.ai for API access or use K2.5 through OpenRouter, Fireworks, or Together AI if you prefer a US based provider. Install Kimi Code from GitHub for a terminal based coding experience.

4

Upgrade or access the API when ready

The free tier is generous for everyday conversations. Upgrade to the paid plan (~$19/month) when you need higher limits for Deep Research, Agent mode, and Agent Swarm. For developers, the K2.5 API at $0.60/$3.00 per million input/output tokens offers frontier level performance at dramatically lower cost than GPT-5.2 or Claude Opus.

Deep dive

Detailed guides with comparisons, tips, and visuals for each feature.

Get notified about updates

We'll email you when this tool's pricing or features change.

Last updated: 2026-02-20