# Fireworks AI Stack

| Layer | Product | Role |
|-------|---------|------|
| **L4** | Blekline | Mask prompts, govern tool calls, audit |
| **L2** | Fireworks AI | Model API / routing |

## Wiring

1. Point SDK `base_url` at `https://app.blekline.com/api/ingress/v1`
2. Set server `OPENAI_API_BASE` to: `https://api.fireworks.ai/inference/v1`
3. Provide upstream API key or IAM credentials on the sidecar (BYOK)

```bash
export BLEKLINE_WORKSPACE_TOKEN=blw_...
export OPENAI_API_BASE=https://api.fireworks.ai/inference/v1
```

**Flow:** Agent/SDK → L4 Blekline ingress → L2 Fireworks AI → model response.

See [Fireworks AI documentation](https://docs.fireworks.ai/).

---

**Next steps:** [Model providers hub](/docs/integrations/model-providers) · [Ingress proxy](/docs/api/ingress-proxy) · [Deployment](/docs/enterprise/deployment) · [Open workspace](https://app.blekline.com)