telnyx-openclaw-embeddings

Telnyx embedding provider for OpenClaw memory search.

Registers Telnyx as a first-class memory embedding provider using the OpenAI-compatible Telnyx AI API.

Features

• Provider ID: telnyx
• Default model: thenlper/gte-large (1024 dimensions)
• Supported models: thenlper/gte-large, intfloat/multilingual-e5-large, Qwen/Qwen3-Embedding-8B
• Auto-select priority: 25
• Auth: TELNYX_API_KEY env var or provider config

Install

openclaw plugins install team-telnyx/telnyx-openclaw-embeddings

Configuration

Minimal (env var)

export TELNYX_API_KEY=KEY...

The provider will be auto-selected when no higher-priority provider is available.

Explicit provider selection

In openclaw.json:

{
  "agents": {
    "defaults": {
      "memorySearch": {
        "provider": "telnyx"
      }
    }
  }
}

Custom model

{
  "agents": {
    "defaults": {
      "memorySearch": {
        "provider": "telnyx",
        "model": "intfloat/multilingual-e5-large"
      }
    }
  }
}

E2E Test

export TELNYX_API_KEY=KEY...
node e2e-test.mjs

Unit Tests

npm install
npm test

Architecture

Uses the same resolveRemoteEmbeddingClient + fetchRemoteEmbeddingVectors infrastructure as the built-in OpenAI provider, since Telnyx exposes an OpenAI-compatible embeddings endpoint at https://api.telnyx.com/v2/ai/openai.

Models