For AI Coding Agents

Ship code faster with
the best coding models

Plug Kyma into Cline, OpenCode, or any coding agent. One API key, active open models, OpenAI-compatible. Free to start.

Top Coding Models

MiniMax M2.5 (80.2% SWE-bench), Qwen 3.6 Plus, Qwen 3 Coder — the best models for autonomous coding.

Fast & Reliable

All providers on paid tier. Auto-fallback when one is busy. Smart routing picks the fastest healthy provider.

OpenAI-Compatible

Drop-in replacement. Works with any tool that supports OpenAI API — Cline, OpenCode, Kilo Code, LangChain, CrewAI.

Best Models for Coding

Ranked by real-world coding benchmarks and developer feedback.

1
MiniMax M2.5#1 Coding Model

Top agentic coding model globally. Best for complex multi-file changes and debugging.

minimax-m2.5

80.2% SWE-bench

2
Qwen 3.6 Plus#1 on Kyma

Alibaba's flagship. 131K context with strong reasoning and multilingual performance. Great for repo-level understanding.

qwen-3.6-plus

78.8% SWE-bench

3
Qwen 3 CoderCode Specialist

Optimized specifically for code generation, debugging, and agentic tool use.

qwen-3-coder

Purpose-built

4
Kimi K2.5Tool Use

Moonshot's active agentic model for autonomous coding with function calling and tool use.

kimi-k2.5

Best agentic

5
Gemma 4 31BBest Value

Google's open model. Runs FREE on Google infra — lowest cost per token on Kyma.

gemma-4-31b

Lowest cost

Setup in 30 Seconds

1. Get your free API key2. Pick your tool below 3. Paste the config.

🔧Cline (VS Code)
Settings → Cline → API Provider: OpenAI Compatible

Base URL: https://kymaapi.com/v1
API Key:  ky-xxxxx (your Kyma key)
Model ID: minimax-m2.5
💻OpenCode (CLI)
opencode config set provider openai-compatible
opencode config set apiKey ky-xxxxx
opencode config set baseURL https://kymaapi.com/v1
opencode config set model qwen-3.6-plus
Kilo Code (VS Code)
Settings → Kilo Code → Provider: OpenAI Compatible

Base URL: https://kymaapi.com/v1
API Key:  ky-xxxxx (your Kyma key)
Model ID: qwen-3-coder
🖱️Cursor
Settings → Models → OpenAI API Key

Base URL: https://kymaapi.com/v1
API Key:  ky-xxxxx (your Kyma key)
Model:    qwen-3.6-plus
🐍Python / OpenAI SDK
from openai import OpenAI

client = OpenAI(
    base_url="https://kymaapi.com/v1",
    api_key="ky-xxxxx"  # your Kyma key
)

response = client.chat.completions.create(
    model="minimax-m2.5",
    messages=[{"role": "user", "content": "Fix this bug..."}]
)
📦Node.js / TypeScript
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://kymaapi.com/v1",
  apiKey: "ky-xxxxx", // your Kyma key
});

const res = await client.chat.completions.create({
  model: "qwen-3.6-plus",
  messages: [{ role: "user", content: "Refactor this code..." }],
});

Not Sure Which Model?

Use our recommendation API to find the best model for your usecase:

GET https://kymaapi.com/v1/models/recommend?usecase=coding
GET https://kymaapi.com/v1/models/recommend?usecase=sql
GET https://kymaapi.com/v1/models/recommend?usecase=fast
GET https://kymaapi.com/v1/models/recommend?usecase=reasoning
GET https://kymaapi.com/v1/models/recommend?usecase=cheap

Start coding with the best models — free

$0.50 free credits on signup. No credit card required.

Get Free API Key