Available Now in Kilo Code

MiniMax M2.7 is here

#5 on PinchBench. $0.30/M input. Available now in Kilo.

86.2%
PinchBench (Top 5 of 50)
$0.30
Per million input tokens
47%
Kilo Bench (89 tasks)

Why M2.7 matters

Near-Opus performance at a fraction of the cost

Near-Opus performance at a fraction of the cost

3.7-point jump from MiniMax M2.5 — moved MiniMax from mid-pack to top tier

Reads deeply before writing — catches things other models miss

Solves tasks no other model can (unique solve on SPARQL reasoning)

Self-evolving model: ran 100+ rounds of autonomous scaffold optimization

“These models aren't interchangeable. They're complementary.”

PinchBench — Top 5

50 models evaluated

1. Opus 4.6
87.4%
2. GLM-5
86.4%
3. GPT-5.4
86.4%
4. M2.7
86.2%
5. Qwen3.5-plus
85.8%
View PinchBench →

Use it everywhere in Kilo

Pick the surface that fits your workflow

VS Code & JetBrains

Select M2.7 from 500+ models in your editor. Switch models per task.

Install Extension →

Kilo CLI

Run M2.7 from your terminal. Full agentic coding, no browser needed.

Get CLI →

KiloClaw (Hosted OpenClaw)

Fully managed cloud agent. 2 clicks, 60 seconds to start. No setup.

Launch KiloClaw →

Trusted by developers at the world's most innovative companies

Why Kilo Code

Open source, no vendor lock-in, zero markup

Open Source

Apache 2.0. Inspect, modify, self-host. No black boxes.

500+ Models

Not locked into one provider. Switch models per task, per project.

BYOK (Bring Your Own Key)

Use your own API keys. Direct provider pricing. No middleman.

$0/month Base Price

Free to start. No subscription required. Ever.

Pay As You Go

Only pay for tokens you use. Zero markup on provider prices.

M2.7 at Provider Cost

$0.30/M input, $1.20/M output. Same as MiniMax direct.

When to use M2.7

With 500+ models in Kilo, pick the right model for the job

Use M2.7 when

  • Complex refactors spanning multiple files
  • Codebase-wide changes needing deep context
  • Tasks where thoroughness > speed
  • Budget-conscious work needing near-frontier quality

Consider lighter models when

  • Quick iteration cycles
  • Well-scoped, time-sensitive edits
  • Simple file changes or boilerplate

Bottom line: With 500+ models in Kilo, you're never locked in.

Agentic Engineering

Glide through your workflow with a mode for every step

Ask mode

A knowledgeable technical assistant focused on answering questions without changing your codebase

Frequently Asked Questions

Everything you need to know about using MiniMax M2.7 with Kilo Code

Compare Kilo Code with Other Tools

See how Kilo Code stacks up against other AI coding assistants

Kilo Code vs Cursor

The open-source agentic platform inside your existing IDE vs the standalone AI-first code editor. 500+ models, zero markup, no editor switch required.

Kilo Code vs GitHub Copilot

Multi-model, multi-mode agentic coding vs a single-vendor autocomplete from GitHub.

Kilo Code vs Windsurf

500+ models at exact provider rates. No credit system. Full BYOK on all plans. Open source.

Kilo Code vs Claude Code

Open-source, multi-model CLI + IDE agent with inline autocomplete vs Anthropic's Claude-only terminal-first coding agent.

Kilo Code vs Roomote

Roomote is the new product from the team behind Roo Code after its May 2026 shutdown. Compare it with Kilo Code — the proven, actively maintained open-source AI coding agent.

Kilo Code vs Roo Code

Roo Code archives May 15, 2026 — compare it with Kilo Code and migrate in minutes.

Kilo Code vs Cline

Kilo Code bundles Cline-style autonomy plus Orchestrator, Architect, Debug, and Code modes.

Kilo Code vs Tabnine

Full agentic coding across 500+ models vs Tabnine's enterprise-only autocomplete.

Kilo Code vs Augment Code

Open-source, BYOK-everywhere agent platform vs Augment Code's closed proprietary stack.

Kilo Code vs Lovable

Kilo Code plugs into your real IDE and codebase — Lovable is a hosted AI app builder optimized for greenfield prototyping.

Kilo Code vs Replit

Use Kilo inside your existing editor and infrastructure vs Replit's hosted browser IDE.

Kilo Code vs Warp

Kilo Code works inside VS Code and JetBrains as an AI coding agent. Warp is a standalone terminal-first Agentic Development Environment. Same agent power, different home.

Kilo Code vs Amp Code

Amp Code is shutting down its VS Code extension — compare it with Kilo Code

Kilo Code for JetBrains IDEs

AI coding assistant for IntelliJ, PhpStorm, WebStorm, and Rider

Kilo CLI vs Claude Code

Compare Kilo CLI with Claude Code for terminal-first AI coding

Kilo CLI vs Aider

Compare model choice, workflows, and pricing for command-line agents

Kilo CLI vs Codex CLI

See how Kilo CLI compares with OpenAI Codex CLI

Kilo CLI vs Gemini CLI

Compare Google Gemini CLI with Kilo CLI and 500+ models

Kilo Autocomplete vs Cursor Tab

Feature-by-feature comparison for AI code completions

Kilo Autocomplete vs GitHub Copilot

Compare autocomplete quality, pricing, and IDE support

Kilo Autocomplete vs Codeium

Compare Kilo Code autocomplete with Codeium completions

Kilo Code Reviews vs Greptile

Compare AI code review workflows, model choice, and pricing

Graphite vs Kilo Code Reviews

Compare Graphite PR workflows with Kilo Code Reviews

Best AI Coding Assistant

Compare top AI coding assistants and when to choose each one

Start using M2.7 in under 60 seconds.

Join 3M+ developers using Kilo Code. Near-Opus performance at $0.30/M input.

#5 PinchBench$0.30/M inputZero markup