Direct answer
The 2026 African developer stack is one API key over a small set of pinned routes, one prepaid USD wallet, and a prompt-set eval you trust more than any leaderboard. Route support calls through deepseek-v4.1-flash-1017, code through glm-5.3-flash or qwen3.8-27b, and retrieval through fikra-embed; pay once at checkout, in Kenya with M-Pesa or Airtel Money, elsewhere by USD card.
Live numbers as of 2026-11-14
These figures come from the public stats snapshot generated on 2026-11-14 from backend ledgers. They show volume and reliability only; they do not prove any single route wins your workload. Run your own 50 prompts before you move budget.
| Metric | Value | What it tells you |
|---|---|---|
| All-time requests | 126,617 | 109 days since launch |
| All-time tokens processed | 5,460,329,009 | 2,299,586,240 served from cache |
| Requests, last 30 days | 116,189 | 5,437,162,745 tokens across 82 active routes |
| Registered developers (all time) | 424 | Self-serve accounts, prepaid |
| Live model routes | 31 | Catalogue at /models; machine list at /v1/models |
| DeepSeek V4.1 Flash 1017, last 30 days | 63,202 | cache-hit rate 12.94% |
| DeepSeek V4.1 Flash Economy (free), last 30 days | 8,741 | cache-hit rate 97.36%; free window, not an SLA |
| 30-day uptime | 98.95% | 88,390 health checks; status currently reports attention, re-check /status before production |
What the routes cost right now
Prices are USD per 1M tokens, taken from the live registry snapshot dated 2026-11-14. The KES columns use the checkout rate of KES 133 = $1; your bank or card issuer may add its own conversion margin on top.
The catalogue is authoritative at /v1/models (the OpenAI-compatible listing) and at /models (the public card with context windows and rate limits). Check both before a production commitment; prices move with provider terms.
See the live list from your terminal: curl -s https://api.fikraapi.co.ke/v1/models -H "Authorization: Bearer YOUR_FIKRA_KEY"
| Model ID | Input / 1M | Cached input / 1M | Output / 1M | Input KES / 1M | Output KES / 1M |
|---|---|---|---|---|---|
| deepseek-v4.1-flash-1017 | $0.05 | $0.0075 | $0.15 | KES 6.65 | KES 19.95 |
| deepseek-v4.1-flash | $0.05 | $0.0075 | $0.15 | KES 6.65 | KES 19.95 |
| qwen3.8-27b | $0.045 | $0.045 | $0.35 | KES 5.98 | KES 46.55 |
| glm-5.3-flash | $0.1875 | $0.0375 | $0.625 | KES 24.94 | KES 83.12 |
| glm-5.3 | $1.225 | $0.225 | $3.85 | KES 162.93 | KES 512.05 |
| kimi-k3 | $2.4375 | $0.24375 | $12.1875 | KES 324.19 | KES 1,620.94 |
The four layers and what belongs in each
Keep the stack to four layers: payments and wallet, routing and keys, inference routes, and evaluation. Everything else is a library choice, not an infrastructure decision.
| Layer | What you need | Where it lives |
|---|---|---|
| Wallet and checkout | Prepaid USD balance, KES 133 = $1 in Kenya | /dashboard checkout |
| Keys and routing | fk_live_ keys per environment, one base_url | https://api.fikraapi.co.ke/v1 |
| Inference routes | Pinned DeepSeek, Qwen, GLM, Kimi, Gemini-family and media routes | /v1/models and /models |
| Evaluation | Your 50-prompt set, accepted-task scoring | Your repo; platform telemetry at /stats |
A workable default stack for a two-person team
- Support drafts and summarisation: deepseek-v4.1-flash-1017 at $0.05 input and $0.15 output per 1M tokens.
- Code review and refactors: glm-5.3-flash at $0.1875/$0.625, or qwen3.8-27b at $0.045/$0.35 for structured output.
- Retrieval over your own corpus: fikra-embed at $0.05 per 1M input tokens, no output token charge.
- One escalation route for the hard 10% of tickets; everything else stays on the default route and the wallet stays flat.
- Budget guard: Lite's $0.50 daily window as a circuit breaker, Pro when two keys are needed, Max when seats are shared.
What breaks teams at this layer
- Opening five provider accounts and bleeding FX margin on five small invoices instead of one top-up.
- Skipping the eval and discovering a regression after the bill landed.
- Pinning nothing, then wondering why a route ID you benchmarked no longer matches the bill.
- Assuming mobile money works outside Kenya; it does not at this checkout.
How to start in ten minutes
- Create the account at /account/signup and verify the email address.
- Copy the fk_live_ key from the dashboard; it is shown once.
- Fund the wallet with M-Pesa, Airtel Money or card in Kenya; elsewhere in Africa pay by USD card. Minimum top-up is $5.
- Set base_url to https://api.fikraapi.co.ke/v1 in your OpenAI client and model to deepseek-v4.1-flash-1017.
- Run 50 prompts from your own workload, then check /stats and the dashboard before scaling spend.
- Cap the key with a budget; expand only after cost per accepted task is where you need it.
curl https://api.fikraapi.co.ke/v1/chat/completions \
-H "Authorization: Bearer YOUR_FIKRA_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"deepseek-v4.1-flash-1017","messages":[{"role":"user","content":"Summarise this ticket in two lines."}]}'
Questions buyers ask
Is this stack production-proven in Africa?
Platform-wide, Fikra processed 126,617 requests over the last 109 days, with 116,189 requests in the last 30 days. Your workload still needs its own 50-prompt check.
Which route starts the stack?
deepseek-v4.1-flash-1017 for text and code. Confirm it is live at /v1/models before pinning it in production.
Do I need a GPU anywhere?
No. Fikra is an API gateway, not a GPU cloud; it routes to pinned third-party models and bills per token.
Is this page legal or procurement advice?
No. It is operating guidance. Verify route eligibility, billing and provider terms at /v1/models and in the provider's own documentation before signing anything.
Does the catalogue list every model in the world?
No. It is a curated set of pinned routes. Availability depends on capacity, licensing and provider terms, and routes are added to the public catalogue as they clear review.