OpenClawSkills
GitHub
Providers • TutorialHeader.readTime

Synthetic

Use Synthetic's Anthropic-compatible API in OpenClaw

Synthetic exposes Anthropic-compatible endpoints. OpenClaw registers it as

'<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'synthetic'</code>' provider and uses Anthropic Messages API.

Tutorial.step

Quick setup

1. Set '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'SYNTHETIC_API_KEY'</code>' (or run wizard below).

2. Run onboarding:

Bash
openclaw onboard --auth-choice synthetic-api-key

The default model is set to:

Terminal
synthetic/hf:MiniMaxAI/MiniMax-M2.1
Tutorial.step

Config example

Json5
{
  env: { SYNTHETIC_API_KEY: "sk-..." },
  agents: {
    defaults: {
      model: { primary: "synthetic/hf:MiniMaxAI/MiniMax-M2.1" },
      models: { "synthetic/hf:MiniMaxAI/MiniMax-M2.1": { alias: "MiniMax M2.1" } },
    },
  },
  models: {
    mode: "merge",
    providers: {
      synthetic: {
        baseUrl: "https://api.synthetic.new/anthropic",
        apiKey: "${SYNTHETIC_API_KEY}",
        api: "anthropic-messages",
        models: [
          {
            id: "hf:MiniMaxAI/MiniMax-M2.1",
            name: "MiniMax M2.1",
            reasoning: false,
            input: ["text"],
            cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
            contextWindow: 192000,
            maxTokens: 65536,
          },
        ],
      },
    },
  },
}

Note: OpenClaw's Anthropic client appends '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'/v1'</code>' to the base URL, so use

'<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'https://api.synthetic.new/anthropic'</code>' (not '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'/anthropic/v1'</code>'). If Synthetic changes

its base URL, override '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'models.providers.synthetic.baseUrl'</code>'.

Tutorial.step

Model catalog

All models below use cost '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'0'</code>' (input/output/cache).

| Model ID | Context window | Max tokens | Reasoning | Input |

| | -- | | |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:MiniMaxAI/MiniMax-M2.1'</code>' | 192000 | 65536 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:moonshotai/Kimi-K2-Thinking'</code>' | 256000 | 8192 | true | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:zai-org/GLM-4.7'</code>' | 198000 | 128000 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:deepseek-ai/DeepSeek-R1-0528'</code>' | 128000 | 8192 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:deepseek-ai/DeepSeek-V3-0324'</code>' | 128000 | 8192 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:deepseek-ai/DeepSeek-V3.1'</code>' | 128000 | 8192 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:deepseek-ai/DeepSeek-V3.1-Terminus'</code>' | 128000 | 8192 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:deepseek-ai/DeepSeek-V3.2'</code>' | 159000 | 8192 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:meta-llama/Llama-3.3-70B-Instruct'</code>' | 128000 | 8192 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8'</code>' | 524000 | 8192 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:moonshotai/Kimi-K2-Instruct-0905'</code>' | 256000 | 8192 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:openai/gpt-oss-120b'</code>' | 128000 | 8192 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:Qwen/Qwen3-235B-A22B-Instruct-2507'</code>' | 256000 | 8192 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:Qwen/Qwen3-Coder-480B-A35B-Instruct'</code>' | 256000 | 8192 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:Qwen/Qwen3-VL-235B-A22B-Instruct'</code>' | 250000 | 8192 | false | text + image |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:zai-org/GLM-4.5'</code>' | 128000 | 128000 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:zai-org/GLM-4.6'</code>' | 198000 | 128000 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:deepseek-ai/DeepSeek-V3'</code>' | 128000 | 8192 | false | text |

| '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'hf:Qwen/Qwen3-235B-A22B-Thinking-2507'</code>' | 256000 | 8192 | true | text |

Tutorial.step

Notes

- Model refs use '<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'synthetic/&lt;modelId&gt;'</code>'.

- If you enable a model allowlist ('<code className="bg-white/10 px-1.5 py-0.5 rounded text-emerald-300 text-sm">'agents.defaults.models'</code>'), add every model you

plan to use.

- See '<a href="/concepts/model-providers" className="text-emerald-400 hover:text-emerald-300 transition-colors">'Model providers'</a>' for provider rules.