# AI model access for OpenClaw

Markdown version of https://kymaapi.com/for/openclaw, for agents and crawlers. Same content as the HTML page.

- Generated: 2026-10-18T19:28:44.827Z
- HTML page: https://kymaapi.com/for/openclaw
- Model catalog: https://kymaapi.com/models.md
- Pricing: https://kymaapi.com/pricing.md

For OpenClaw

Connect OpenClaw to Kyma API for model choice and per-model prompt caching. $0.50 signup credit for eligible free-tier models after email verification.

- [Create account](https://kymaapi.com/signup)
- [Setup Guide](https://kymaapi.com/for/openclaw#setup)

### One account for model access

Use one key and balance for compatible models. Compare current performance on the rankings page.

### Per-model cached-input rates

OpenClaw's repeated context gets cached automatically, and the repeated part bills at that model's own cached-input rate.

### Automatic fallback

For eligible failures, Kyma API tries another compatible route when available. Requests can still fail.

## Models for OpenClaw

Examples with tool calling and long context. Check current pricing and measured uptime before choosing.

- 1 · Kimi K2.6 · Vision input · Supports reasoning, tool use, vision input. Compare current prices and measured uptime on the model page. · `kimi-k2.6` · 262K ctx
- 2 · Qwen 3.6 Plus · Vision input · Supports reasoning, tool use, vision input. Compare current prices and measured uptime on the model page. · `qwen-3.6-plus` · 1M ctx
- 3 · DeepSeek V4 Flash · Tool use · Supports reasoning, tool use. Compare current prices and measured uptime on the model page. · `deepseek-v4-flash` · 1M ctx
- 4 · Gemma 4 31B · Vision input · Supports reasoning, tool use, vision input. Compare current prices and measured uptime on the model page. · `gemma-4-31b` · 128K ctx
- 5 · Llama 3.3 70B · Tool use · Supports tool use. Compare current prices and measured uptime on the model page. · `llama-3.3-70b` · 128K ctx

## Setup guide

1. [Create account](https://kymaapi.com/signup) → 2. Add the config below → 3. Start chatting.

**OpenClaw Setup**

```
// ~/.openclaw/openclaw.json
{
  "models": {
    "providers": [
      {
        "name": "Kyma",
        "api": "openai-completions",
        "baseUrl": "https://kymaapi.com/v1",
        "apiKey": "kyma-xxxxx",
        "models": [
          "kimi-k2.6",
          "qwen-3.6-plus",
          "deepseek-v4-flash",
          "gemma-4-31b",
          "llama-3.3-70b"
        ]
      }
    ],
    "default": "kimi-k2.6"
  }
}
```

**Quick Start (3 Steps)**

```
Step 1: Create account
  → Sign up at kymaapi.com

Step 2: Add Kyma to OpenClaw
  → Edit ~/.openclaw/openclaw.json
  → Add the Kyma provider config (see left)

Step 3: Start chatting
  → OpenClaw will use Kyma's models automatically
  → Type in any messaging app connected to OpenClaw

Tip: Use "gemma-4-31b" for cheap high-volume inference
     Use "kimi-k2.6" for best agentic quality
```

## Create account

$0.50 signup credit for eligible free-tier models after email verification. No card required.

[Create account](https://kymaapi.com/signup)

## Explore more

- [Kimi K2.6](https://kymaapi.com/models/kimi-k2.6)
- [Qwen 3.6 Plus](https://kymaapi.com/models/qwen-3.6-plus)
- [DeepSeek V4 Flash](https://kymaapi.com/models/deepseek-v4-flash)
- [Gemma 4 31B](https://kymaapi.com/models/gemma-4-31b)
- [Llama 3.3 70B](https://kymaapi.com/models/llama-3.3-70b)
- [Compare models](https://kymaapi.com/compare)
- [Live model rankings](https://kymaapi.com/rankings)
- [OpenClaw setup guide](https://docs.kymaapi.com/guides/openclaw)