diff --git a/content/docs/configuration/librechat_yaml/ai_endpoints/fikra.mdx b/content/docs/configuration/librechat_yaml/ai_endpoints/fikra.mdx new file mode 100644 index 000000000..e76fa12fb --- /dev/null +++ b/content/docs/configuration/librechat_yaml/ai_endpoints/fikra.mdx @@ -0,0 +1,29 @@ +--- +title: Fikra API +icon: Globe +description: Configure Fikra API as a custom endpoint in LibreChat. +--- + +Fikra API is an OpenAI-compatible inference service serving open models (DeepSeek, GLM, Kimi, Qwen) behind one endpoint. Billing is per token from a prepaid wallet, with M-Pesa, MTN MoMo, Airtel Money and card payments accepted. New accounts start with a $1 signup credit. + +## Configuration + +Add the endpoint under `endpoints.custom` in your `librechat.yaml`: + +```yaml filename="librechat.yaml" + - name: 'Fikra API' + apiKey: '${FIKRA_API_KEY}' + baseURL: 'https://api.fikraapi.co.ke/v1' + models: + default: ['deepseek-v4.1-flash'] + fetch: true + titleConvo: true + titleModel: 'current_model' + modelDisplayLabel: 'Fikra' +``` + +## Notes + +- With `fetch: true`, LibreChat retrieves the live model list from the endpoint. Model ids such as `deepseek-v4.1-flash`, `glm-5.3-flash` and `kimi-k3` work as-is. The full list lives in the [model catalog](https://fikraapi.co.ke/models). +- Pricing is per token. Live rates are published at [fikraapi.co.ke/pricing](https://fikraapi.co.ke/pricing). +- `fikra-auto` routes each request to a suitable model automatically. diff --git a/content/docs/configuration/librechat_yaml/ai_endpoints/meta.json b/content/docs/configuration/librechat_yaml/ai_endpoints/meta.json index c46fe3dfd..540f43632 100644 --- a/content/docs/configuration/librechat_yaml/ai_endpoints/meta.json +++ b/content/docs/configuration/librechat_yaml/ai_endpoints/meta.json @@ -9,6 +9,7 @@ "cohere", "databricks", "deepseek", + "fikra", "fireworks", "groq", "helicone", @@ -31,4 +32,4 @@ "vultrcloudinference", "xai" ] -} +} \ No newline at end of file