muapi Logomuapi Logo
muapi Logomuapi Logo

Moonshot Kimi K3 MoE LLM: AI Large Language Models

Moonshot Kimi K3 is a flagship 2.8T Mixture-of-Experts (MoE) LLM with a 1M token context window for long-context reasoning, coding, and complex agent workflows.

Interactive model controls

Moonshot Kimi K3 is a flagship 2.8T Mixture-of-Experts (MoE) LLM with a 1M token context window, designed for long-context reasoning, coding, and complex agent workflows. Token-based pricing: $3.00/M input tokens and $15.00/M output tokens. Two endpoints: standard async (/kimi-k3) and live streaming (/kimi-k3/stream) via SSE.

📝

Overview

About this model

Moonshot Kimi K3 is a flagship 2.8T Mixture-of-Experts (MoE) LLM featuring a massive 1M token context window. Designed for deep reasoning, multi-turn coding, and complex agentic task execution.

1Long-context document analysis and repository-level code comprehension
2Complex multi-step reasoning and algorithmic problem solving
3Autonomous AI agent workflows and interactive conversational agents
💰

Pricing & Value

Cost analysis

muapiapp$0.0001 per request

Token-based pricing ($3.00/M input tokens, $15.00/M output tokens).

** Competitor pricing is estimated based on similar model architectures and usage tiers.

⚙️

Technical Details

Configuration schema

Promptstring

The user message or instruction for the model.

Default ValueExplain the architecture of Mixture-of-Experts models.
Image URLstring

Optional image URL to include as multimodal input.

Default Valueundefined
System Promptstring

Optional system-level instruction to guide model behavior.

Default ValueYou are an expert AI researcher and coding assistant.
Temperaturenumber

Controls output randomness (0.0 to 2.0). Lower values are more deterministic.

Default Value1
Max Tokensinteger

Maximum number of tokens to generate.

Default Value4096
Top Pnumber

Nucleus sampling parameter (0.0 to 1.0).

Default Value1
Frequency Penaltynumber

Frequency penalty (-2.0 to 2.0). Reduces likelihood of repeating words.

Default Value0
Presence Penaltynumber

Presence penalty (-2.0 to 2.0). Increases likelihood of new topics.

Default Value0
📖

Implementation Guide

Developer documentation

How to Use Moonshot Kimi K3

  1. Provide Prompt: Send a prompt or conversation history to /api/v1/kimi-k3 or stream via /api/v1/kimi-k3/stream.
  2. Configure Parameters: Optionally set system_prompt, temperature, max_tokens, top_p, frequency_penalty, and presence_penalty.
  3. Multimodal Input: Optionally include an image_url for multimodal processing.
  4. Receive Response: Get generated text responses with reasoning and code completion.
❓

Common Questions

Frequently asked

What is the context window size of Moonshot Kimi K3?

Moonshot Kimi K3 supports up to a 1,000,000 (1M) token context window.

Does Moonshot Kimi K3 support streaming responses?

Yes, live real-time token streaming is supported via Server-Sent Events (SSE) at `/api/v1/kimi-k3/stream`.