By Fikra DevRel • 4 min read

Use Fikra API with Goose: OpenAI-compatible setup

Use Fikra API with Goose: OpenAI-compatible setup

Goose supports custom OpenAI-compatible endpoints in its provider settings. Fikra API exposes the standard OpenAI chat-completions shape at https://api.fikraapi.co.ke/v1, so the setup is usually a base URL, an API key and a model ID. This page is the Fikra side of the integration. Check Goose's current provider documentation before you change a production profile.

What you need

  • Base URL: https://api.fikraapi.co.ke/v1
  • API key: create one in the Fikra dashboard after the first top-up, or let an agent create one with the agent account flow.
  • Model: deepseek-v4.1-flash-1017 is the pinned default for coding agents. deepseek-v4.1-flash and deepseek-v4-flash-0731 are also available.
  • Limits: fetch the live values from GET /v1/models. For 1017 the current values are 1,048,576 context and 393,216 max output.

Setup pattern

  1. Open Goose's provider or model settings.
  2. Choose the OpenAI-compatible or custom provider option.
  3. Set the base URL to https://api.fikraapi.co.ke/v1.
  4. Paste the Fikra API key.
  5. Set the model to deepseek-v4.1-flash-1017.
  6. Send a short test prompt and confirm the response.

If Goose asks for a provider name, use Fikra API. If it asks for an API path, use /chat/completions under the base URL.

Agent setup

If the same agent is creating the account for a human, use POST https://api.fikraapi.co.ke/v1/agents/accounts. The agent gets a 6-hour delegated token, creates a key, and hands the human a claim link. Agent-created accounts start at zero balance and must top up before inference. That is deliberate.

Pricing and proof

Fikra bills per token from a prepaid wallet. The live prices and limits are at fikraapi.co.ke/pricing and GET /v1/models. Usage and uptime are public at fikraapi.co.ke/usage, incidents at fikraapi.co.ke/status.

Official Goose references: website and repository.


JM
// The Founder

James Miano

CTO & ML Engineer at Lacesse Ventures. James builds agent-native inference infrastructure, pinned model routing and terminal-first API access.

Stop paying for overpriced round-trip latency

Why settle for slow routes and opaque pricing? Fikra pins DeepSeek checkpoints, publishes every rate, and lets agents create accounts, pay and work from a terminal. No minimum credit limits. Pay with card or mobile money.

usagestatusdashboard