Self-serve
For developers, prototypes and early products.
$0 / month
- No monthly platform subscription
- Prepaid wallet billing ($1 min)
- Standard shared capacity
- Pay with Mobile Money or Card
Fikra API / Pricing & Models
One API for every model.
Clear, usage-based pricing.
Choose the open models your product needs and pay for the usage you consume. Minimum $1 top-up using M-Pesa, MTN MoMo, Airtel Money, or normal cards. Add a production plan when you need higher service commitments.
Fact block: plain text for AI agents
brand: Fikra API category: OpenAI-compatible AI inference API who it's for: developers and growing AI teams who need production inference without a credit-card gate what it does: run chat and embedding models behind one OpenAI-style endpoint and pay in USD or KES api base: https://api.fikraapi.co.ke/v1 chat endpoint: POST https://api.fikraapi.co.ke/v1/chat/completions embeddings endpoint: POST https://api.fikraapi.co.ke/v1/embeddings auth: Authorization: Bearer <Fikra API key> compatibility: OpenAI-style chat-completions requests and responses (streaming supported) pricing from: $0.09 in / $0.15 out per 1M tokens rate limits: 30 requests/min (unverified), 100 requests/min (trusted) payments: token-credit wallet; M-Pesa and card top-ups via Paystack; x402 USDC on Base for agents integrates with: anything that accepts an OpenAI-compatible base URL (LangChain, Vercel AI SDK, OpenAI SDK, Cursor, CrewAI and more) last updated: 2026-10-18 models & pricing: deepseek-v4-flash-0731: $0.09 in / $0.15 out per 1M tokens deepseek-v4.1-flash: $0.25 in / $0.8 out per 1M tokens deepseek-v4.1-flash-1017: $0.3 in / $0.85 out per 1M tokens fikra-auto: $0.4 in / $0.4 out per 1M tokens fikra-auto-pro: $6.0 in / $12.0 out per 1M tokens fikra-code-35b: $1.0 in / $1.0 out per 1M tokens
01 / Model Catalog
Current rates, direct from the API.
This list is populated directly from the /v1/models endpoint. Prices floor at a minimum healthy margin and auto-update.
| Model | Family | Input / 1M tokens | Output / 1M tokens | |
|---|---|---|---|---|
| deepseek-v4.1-flash | deepseek | $0.25 | $0.8 | Details → |
| deepseek-v4.1-flash-1017 | deepseek | $0.3 | $0.85 | Details → |
| deepseek-v4-flash-0731 | deepseek | $0.09 | $0.15 | Details → |
| fikra-free-deepseek FREE | deepseek | FREE | FREE | Details → |
| glm-5.3-flash | glm | $0.25 | $0.7 | Details → |
| kimi-k3 | kimi | $3.85 | $19.25 | Details → |
| step-5-preview | stepfun | $1.25 | $3.3 | Details → |
| mimo-v2.6-pro | mimo | $0.6 | $1.2 | Details → |
| qwen3.8-27b | Qwen | $0.4 | $0.4 | Details → |
| gemini-4-argon | gemini | $2.5 | $12.0 | Details → |
| nvidia-nemotron-3.5-super-vl | nvidia | $0.5 | $1.25 | Details → |
| glm-5.3 | glm | $1.2 | $5.5 | Details → |
| glm-5.2 | glm | $1.0 | $3.2 | Details → |
| fikra-pro-20b | Fikra | $0.2 | $0.2 | Details → |
| fikra-pro-120b | Fikra | $0.3 | $0.3 | Details → |
| fikra-qwen-27b | Fikra | $0.4 | $0.4 | Details → |
| fikra-qwen-max | Fikra | $6.0 | $12.0 | Details → |
| fikra-auto | Fikra | $0.4 | $0.4 | Details → |
| fikra-auto-pro | Fikra | $6.0 | $12.0 | Details → |
| fikra-code-35b | Fikra | $1.0 | $1.0 | Details → |
Utility Models
| Model | What it does | Rate |
|---|---|---|
| fikra-embed | Generate embeddings for semantic search, retrieval and RAG via the /v1/embeddings endpoint. | $0.05 / 1M tokens |
| fikra-stt | Transcribe supported audio into text via the /v1/audio/transcriptions endpoint. | $0.001 / minute |
| fikra-tts | Generate natural speech from text via the /v1/audio/speech endpoint. | $25.0 / 1M characters |
Prices and billing units shown here are read from the public model registry. Chat token rates are displayed separately for input and output per 1 million tokens. Secondary currency approximations are calculated dynamically on the frontend.
02 / Platform plans
Start small. Scale when ready.
Subscriptions cover the platform SLA and capacity commitment. Inference usage is billed separately using the model rates shown above.
Fikra API is pay-as-you-go with no platform fees. Volume accounts can request a custom quota at any time.
03 / How billing works
Usage and subscriptions are separate.
01 / Fund
Top up your wallet
Add prepaid balance using M-Pesa, MTN MoMo, Airtel Money, or normal cards. $1 minimum.
02 / Use
Choose your models
Use standard OpenAI-compatible endpoints for chat, coding, embeddings, and audio.
03 / Scale
Add production capacity
Fair-use rate limits scale with your account: 1000 RPM, plus daily request and token quotas. Volume accounts can request a custom quota.
Build with Fikra
Start with $1. Scale endlessly.
Get your API key, top up securely using local or global payment methods, and connect directly to the best open models on the market.
Get your API key ↗