Skip to main content

Agents

Use Voxvey as the model gateway for coding agents and editor assistants that accept Chat Completions, Responses, or Messages API endpoints.

Connection details

SettingValue
OpenAI-compatible base URLhttps://api.voxvey.com/v1
Messages base URLhttps://api.voxvey.com
Chat completions endpointhttps://api.voxvey.com/v1/chat/completions
Responses endpointhttps://api.voxvey.com/v1/responses
Messages endpointhttps://api.voxvey.com/v1/messages
AuthBearer token
Token env varVOXVEY_TOKEN

Set your token before configuring a local agent:

export VOXVEY_TOKEN="<access-token>"

Use the model ID that matches the API surface your client calls.

Client surfaceModel ID
Chat Completionsdeepseek/deepseek-v4-flash
Responsesopenai/gpt-5.5
MessagesUse the model ID returned by /v1/models for the Messages API.

Setup pages

ClientUse when
Voxvey OpenClaw PluginYou want OAuth sign-in, live model discovery, and the packaged Voxvey provider.
OpenClawYou want a custom provider in OpenClaw config.
HermesYou want Hermes to use Voxvey as the main model endpoint.
Claude CodeYou want Claude Code to use Voxvey's Messages API route.
CodexYou want Codex to use Voxvey's Responses-compatible route.
VS CodeYou want Copilot Chat BYOK Custom Endpoint setup.
ZedYou want Zed Agent or Inline Assistant through an OpenAI-compatible provider.

Verify the gateway

Before debugging an editor or agent, verify the token and model directly.

curl https://api.voxvey.com/v1/chat/completions \
-H "Authorization: Bearer $VOXVEY_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek/deepseek-v4-flash",
"messages": [
{"role": "user", "content": "Reply with: ready"}
],
"stream": false
}'

Expected response shape:

{
"id": "chatcmpl_...",
"object": "chat.completion",
"choices": [
{
"message": {
"role": "assistant",
"content": "ready"
}
}
],
"usage": {
"prompt_tokens": 12,
"completion_tokens": 1,
"total_tokens": 13
}
}