Connect OpenClaw to Synthetic via the Anthropic Messages API
Set up Synthetic's Anthropic-compatible endpoints in OpenClaw using the official provider plugin. Covers plugin installation, API key setup, onboarding, and verifying the default model.
Read this when
- You want to use Synthetic as a model provider
- You need a Synthetic API key or base URL setup
Synthetic offers endpoints that work with Anthropic.
OpenClaw delivers access to them via the official @openclaw/synthetic-provider plugin, and it relies on the Anthropic Messages API.
| Property | Value |
|---|---|
| Provider | synthetic |
| Auth | SYNTHETIC_API_KEY |
| API | Anthropic Messages |
| Base URL | https://api.synthetic.new/anthropic |
Getting started
Install the plugin
openclaw plugins install @openclaw/synthetic-provider
If a Gateway is already running, the installation is applied to it automatically. Otherwise it takes effect the next time it starts. See Apply changes and inspect.
Get an API key
Obtain a SYNTHETIC_API_KEY from your Synthetic account, or let onboarding ask you for one.
Run onboarding
openclaw onboard --auth-choice synthetic-api-key
Verify the default model
Onboarding registers synthetic/hf:MiniMaxAI/MiniMax-M3 as the default model.
To confirm it is registered:
openclaw models list --provider synthetic
Warning
OpenClaw's Anthropic client adds
/v1to the base URL on its own, so supplyhttps://api.synthetic.new/anthropic(not/anthropic/v1). Should Synthetic change its base URL, overridemodels.providers.synthetic.baseUrl.
Config example
{
env: { vars: { SYNTHETIC_API_KEY: "sk-..." } },
agents: {
defaults: {
model: { primary: "synthetic/hf:MiniMaxAI/MiniMax-M3" },
models: { "synthetic/hf:MiniMaxAI/MiniMax-M3": { alias: "MiniMax M3" } },
},
},
models: {
mode: "merge",
providers: {
synthetic: {
baseUrl: "https://api.synthetic.new/anthropic",
apiKey: "${SYNTHETIC_API_KEY}",
api: "anthropic-messages",
models: [
{
id: "hf:MiniMaxAI/MiniMax-M3",
name: "MiniMax M3",
reasoning: true,
input: ["text", "image"],
cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
contextWindow: 262144,
maxTokens: 65536,
},
],
},
},
},
}
Model discovery
When a Synthetic credential is present, OpenClaw pulls the current text models from
Synthetic's /openai/v1/models API.
Inference continues to go through the Anthropic Messages API. Models that appear later, small models and syn: aliases included, require no catalog update in OpenClaw.
Context and output limits, image input, reasoning, tool support, and usage-based token prices all come from the live catalog. Those prices are estimates rather than a subscription bill. Consult Synthetic's current model list for availability and for its recommended aliases.
When catalog generation runs offline, or when discovery responses are missing or unusable, the bundled seed models are used. The model you selected is never switched automatically. If you override the inference base URL, OpenClaw bypasses Synthetic's fixed discovery URL, which keeps a proxy credential from being sent to Synthetic.
Tip
Model refs take the form
synthetic/<modelId>. To inspect your configured models, useopenclaw models list --provider synthetic.
Model allowlist
When a model allowlist is enabled (agents.defaults.modelPolicy.allow), every
Synthetic model you intend to use must be added to it. Any model left out of the allowlist is hidden
from the agent.
Base URL override
If Synthetic changes its API endpoint, override the base URL:
{
models: {
providers: {
synthetic: {
baseUrl: "https://new-api.synthetic.new/anthropic",
},
},
},
}
OpenClaw continues to add /v1 automatically.
Related
-
Model providers, Provider rules, model refs, and failover behavior.
-
Configuration reference, Full config schema including provider settings.
-
Synthetic, Synthetic dashboard and API docs.