Anforderungen  |   Konzepte  |   Entwurf  |   Entwicklung  |   Qualitätssicherung  |   Lebenszyklus  |   Steuerung
 
 
 
 


Quelle  synthetic.md   Sprache: unbekannt

 
Spracherkennung für: .md vermutete Sprache: Unknown {[0] [0] [0]} [Methode: Schwerpunktbildung, einfache Gewichte, sechs Dimensionen]

---
summary: "Use Synthetic's Anthropic-compatible API in OpenClaw"
read_when:
  - You want to use Synthetic as a model provider
  - You need a Synthetic API key or base URL setup
title: "Synthetic"
---

[Synthetic](https://synthetic.new) exposes Anthropic-compatible endpoints.
OpenClaw registers it as the `synthetic` provider and uses the Anthropic
Messages API.

| Property | Value                                 |
| -------- | ------------------------------------- |
| Provider | `synthetic`                           |
| Auth     | `SYNTHETIC_API_KEY`                   |
| API      | Anthropic Messages                    |
| Base URL | `https://api.synthetic.new/anthropic` |

## Getting started

<Steps>
  <Step title="Get an API key">
    Obtain a `SYNTHETIC_API_KEY` from your Synthetic account, or let the
    onboarding wizard prompt you for one.
  </Step>
  <Step title="Run onboarding">
    ```bash
    openclaw onboard --auth-choice synthetic-api-key
    ```
  </Step>
  <Step title="Verify the default model">
    After onboarding the default model is set to:
    ```
    synthetic/hf:MiniMaxAI/MiniMax-M2.5
    ```
  </Step>
</Steps>

<Warning>
OpenClaw's Anthropic client appends `/v1` to the base URL automatically, so use
`https://api.synthetic.new/anthropic` (not `/anthropic/v1`). If Synthetic
changes its base URL, override `models.providers.synthetic.baseUrl`.
</Warning>

## Config example

```json5
{
  env: { SYNTHETIC_API_KEY: "sk-..." },
  agents: {
    defaults: {
      model: { primary: "synthetic/hf:MiniMaxAI/MiniMax-M2.5" },
      models: { "synthetic/hf:MiniMaxAI/MiniMax-M2.5": { alias: "MiniMax M2.5" } },
    },
  },
  models: {
    mode: "merge",
    providers: {
      synthetic: {
        baseUrl: "https://api.synthetic.new/anthropic",
        apiKey: "${SYNTHETIC_API_KEY}",
        api: "anthropic-messages",
        models: [
          {
            id: "hf:MiniMaxAI/MiniMax-M2.5",
            name: "MiniMax M2.5",
            reasoning: false,
            input: ["text"],
            cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
            contextWindow: 192000,
            maxTokens: 65536,
          },
        ],
      },
    },
  },
}
```

## Built-in catalog

All Synthetic models use cost `0` (input/output/cache).

| Model ID                                               | Context window | Max tokens | Reasoning | Input        |
| ------------------------------------------------------ | -------------- | ---------- | --------- | ------------ |
| `hf:MiniMaxAI/MiniMax-M2.5`                            | 192,000        | 65,536     | no        | text         |
| `hf:moonshotai/Kimi-K2-Thinking`                       | 256,000        | 8,192      | yes       | text         |
| `hf:zai-org/GLM-4.7`                                   | 198,000        | 128,000    | no        | text         |
| `hf:deepseek-ai/DeepSeek-R1-0528`                      | 128,000        | 8,192      | no        | text         |
| `hf:deepseek-ai/DeepSeek-V3-0324`                      | 128,000        | 8,192      | no        | text         |
| `hf:deepseek-ai/DeepSeek-V3.1`                         | 128,000        | 8,192      | no        | text         |
| `hf:deepseek-ai/DeepSeek-V3.1-Terminus`                | 128,000        | 8,192      | no        | text         |
| `hf:deepseek-ai/DeepSeek-V3.2`                         | 159,000        | 8,192      | no        | text         |
| `hf:meta-llama/Llama-3.3-70B-Instruct`                 | 128,000        | 8,192      | no        | text         |
| `hf:meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8` | 524,000        | 8,192      | no        | text         |
| `hf:moonshotai/Kimi-K2-Instruct-0905`                  | 256,000        | 8,192      | no        | text         |
| `hf:moonshotai/Kimi-K2.5`                              | 256,000        | 8,192      | yes       | text + image |
| `hf:openai/gpt-oss-120b`                               | 128,000        | 8,192      | no        | text         |
| `hf:Qwen/Qwen3-235B-A22B-Instruct-2507`                | 256,000        | 8,192      | no        | text         |
| `hf:Qwen/Qwen3-Coder-480B-A35B-Instruct`               | 256,000        | 8,192      | no        | text         |
| `hf:Qwen/Qwen3-VL-235B-A22B-Instruct`                  | 250,000        | 8,192      | no        | text + image |
| `hf:zai-org/GLM-4.5`                                   | 128,000        | 128,000    | no        | text         |
| `hf:zai-org/GLM-4.6`                                   | 198,000        | 128,000    | no        | text         |
| `hf:zai-org/GLM-5`                                     | 256,000        | 128,000    | yes       | text + image |
| `hf:deepseek-ai/DeepSeek-V3`                           | 128,000        | 8,192      | no        | text         |
| `hf:Qwen/Qwen3-235B-A22B-Thinking-2507`                | 256,000        | 8,192      | yes       | text         |

<Tip>
Model refs use the form `synthetic/<modelId>`. Use
`openclaw models list --provider synthetic` to see all models available on your
account.
</Tip>

<AccordionGroup>
  <Accordion title="Model allowlist">
    If you enable a model allowlist (`agents.defaults.models`), add every
    Synthetic model you plan to use. Models not in the allowlist will be hidden
    from the agent.
  </Accordion>

  <Accordion title="Base URL override">
    If Synthetic changes its API endpoint, override the base URL in your config:

    ```json5
    {
      models: {
        providers: {
          synthetic: {
            baseUrl: "https://new-api.synthetic.new/anthropic",
          },
        },
      },
    }
    ```

    Remember that OpenClaw appends `/v1` automatically.

  </Accordion>
</AccordionGroup>

## Related

<CardGroup cols={2}>
  <Card title="Model selection" href="/concepts/model-providers" icon="layers">
    Provider rules, model refs, and failover behavior.
  </Card>
  <Card title="Configuration reference" href="/gateway/configuration-reference" icon="gear">
    Full config schema including provider settings.
  </Card>
  <Card title="Synthetic" href="https://synthetic.new" icon="arrow-up-right-from-square">
    Synthetic dashboard and API docs.
  </Card>
</CardGroup>

[Dauer der Verarbeitung: 0.15 Sekunden, vorverarbeitet 2026-04-27]

                                                                                                                                                                                                                                                                                                                                                                                                     


Neuigkeiten

     Aktuelles
     Motto des Tages

Software

     Produkte
     Quellcodebibliothek

Aktivitäten

     Artikel über Sicherheit
     Anleitung zur Aktivierung von SSL

Muße

     Gedichte
     Musik
     Bilder

Jenseits des Üblichen ....

Besucherstatistik

Besucherstatistik

Monitoring

Montastic status badge