Skip to main content
Kaino.dev
Discover
Evals
News
Academics
Insights
Kaino.dev

Discover, evaluate, and compare AI tools, models, and agents.

Explore

  • Discover
  • Evaluations
  • News
  • Academics
  • Insights

Community

  • Twitter
  • YouTube
  • Instagram
Privacy PolicyTerms of Service

© 2026 Kaino.dev. All rights reserved.

Version 1.1.0
Kimi K2.7 Code · Discover · Kaino
Discover/MODELS/Kimi K2.7 Code
Kimi K2.7 Code logo

MODELS

Kimi K2.7 Code

by Moonshot AI

Moonshot AIKimimultimodal
Visit WebsiteDocumentationGitHub

Overview

Open-weight, coding-focused Kimi model for long-horizon software engineering, tool use, large-context codebase work, and coding agents.

Details

Kimi K2.7 Code is a Moonshot AI / Kimi coding model described in supplied official and semi-official sources as an open-source or open-weight agentic coding model. The Kimi resource page lists 1T total parameters, 32B active parameters, 256K context, multimodal input, and Kimi API access. The Kimi API quickstart identifies model IDs kimi-k2.7-code and kimi-k2.7-code-highspeed, says the API is OpenAI-compatible at api.moonshot.ai/v1, and documents image/video use, tool calling, thinking-mode constraints, and request parameters. The Hugging Face model card lists the model under a modified MIT license and summarizes its MoE architecture, MoonViT vision encoder, deployment guidance, and API usage examples.

When to Use

Use for coding-agent and long-horizon software engineering workflows that need a coding-focused Kimi model. Use when working with large codebases or tasks that benefit from a 256K context length. Use when your application needs tool calling JSON Mode Partial Mode or multimodal image/video input alongside coding assistance. Use when evaluating open-weight coding models with official API access and Hugging Face model-card materials.

Getting Started

  1. Read the Kimi K2.7 Code quickstart: https://platform.kimi.ai/docs/guide/kimi-k2-7-code-quickstart
  2. Use the documented model IDs kimi-k2.7-code or kimi-k2.7-code-highspeed with the OpenAI-compatible API at api.moonshot.ai/v1.
  3. Review pricing for the coding model: https://platform.kimi.ai/docs/pricing/chat-k27-code
  4. Inspect the Hugging Face model card for license
  5. architecture
  6. deployment guidance
  7. and examples: https://huggingface.co/moonshotai/Kimi-K2.7-Code
  8. Review the MoonshotAI/Kimi-K2 GitHub repository for related project materials: https://github.com/MoonshotAI/Kimi-K2

Key Features

  • •Coding-focused Kimi model for software engineering and agentic coding workflows
  • •Model IDs documented as kimi-k2.7-code and kimi-k2.7-code-highspeed
  • •OpenAI-compatible API documented at api.moonshot.ai/v1
  • •256K context length
  • •Supports text
  • •image
  • •and video input according to Kimi pricing/API documentation
  • •Supports ToolCalls
  • •JSON Mode
  • •Partial Mode
  • •and automatic context caching
  • •Hugging Face model card lists MoE architecture with 1T total parameters and 32B active parameters

Capabilities

  • •code-focused language modeling
  • •software engineering assistance
  • •coding-agent workflows
  • •tool calling
  • •large-context codebase work
  • •multimodal image input
  • •multimodal video input
  • •JSON-mode responses
  • •partial-mode responses
  • •automatic context caching

Last updated Jun 30, 2026