How do you build Swahili AI apps with Plugsky?

Key facts

Swahili text	Latin script, Bantu noun classes and agglutinative verb forms
Code-switching	English mixing is common in technical and business text
Tokenisation	Morpheme-heavy forms split into subwords — measure with the Plugsky token calculator
API compatibility	OpenAI-compatible POST https://api.plugsky.com/v1/chat/completions — chat, streaming, JSON mode and function calling
Models	30+ models behind one endpoint, from free tiers to frontier reasoning
Free tier	Free plan with plugsky-micro and plugsky-lite, no card required
Deployment	Plugsky cloud, your VPC, on-prem and air-gapped options
Product status	Chat, streaming, JSON mode, function calling, embeddings, RAG and agents live; audio, images, batch and fine-tuning coming soon

TL;DR

• Swahili works through the standard chat endpoint — one base_url change.
• Agglutinated forms split into tokens; measure mixed sentences.
• Treat English code-switching as expected, not an error.
• Stem prefixes for search; keep them for display.
• plugsky-embed-multilingual covers Swahili and English retrieval.

How it works, step by step

1. Create a Plugsky API key on the free plan (no card) and set base_url to https://api.plugsky.com/v1.
2. Send a small set of real Swahili prompts to /v1/chat/completions and compare output across two or three models.
3. Count tokens for those prompts with the Plugsky token calculator and set chunk sizes that fit your model context.
4. Normalise text before indexing or prompting: standardise spelling variants and keep English loanwords as written.
5. For RAG, embed with plugsky-embed-multilingual and test cross-language queries alongside Swahili-only queries.
6. Score candidate models on a Swahili gold set with native-speaker review, then cut production traffic over.

Try it yourself

Open the token calculator →

How Plugsky handles Swahili text

Swahili uses the Latin alphabet and reads left to right, with Bantu noun-class prefixes and agglutinative verb forms.

Standard Kiswahili (Tanzania) and Kenyan usage differ slightly, and English code-switching is common in technical and business text.

Evaluate with a Swahili speaker on pure and code-switched samples; check noun-class agreement, which is the most common error and easy to spot.

Tokenisation and cost in Swahili

Swahili words carry prefixes for noun class, tense and person, so tokens often split at morpheme boundaries; code-switched English words tokenise like English, making mixed sentences uneven.

• Keep noun-class prefixes intact for display; stem them for search.
• Handle English code-switching as part of the same corpus.
• Test Tanzanian and Kenyan spelling and vocabulary variants.
• Count tokens on mixed Swahili and English text, not pure Swahili only.

Swahili retrieval and RAG

Index Swahili documents with their English loanwords and test queries that mix both languages, because users rarely stay in one language.

• Use plugsky-embed-multilingual for Swahili and English corpora.
• Keep code-switched sentences whole when chunking.
• Add stemmed keys for agglutinated forms.
• Evaluate pure and mixed-language queries.

Code example: a Swahili request

Point your existing OpenAI client at https://api.plugsky.com/v1 and pass Swahili text in the content field — no language flag and no separate endpoint. Streaming, JSON mode and function calling keep the same request shapes.

client = OpenAI(base_url="https://api.plugsky.com/v1", api_key=os.environ["PLUGSKY_API_KEY"])

client.chat.completions.create(model="plugsky-pro", messages=[{"role": "user", "content": "Fanya muhtasari wa mkataba huu kwa Kiswahili, pointi tatu."}])

Start on the free plan with plugsky-micro and plugsky-lite, then compare paid models on a Swahili gold set before cutover. See the docs for request details.

For production, log the model name and your normalisation settings with each request, and re-run the Swahili gold set whenever either changes — language quality regressions usually come from prompt or preprocessing drift, not from the model alone.

Honest comparison

Capability	Plugsky	Swahili workflow today	Building in-house
API compatibility	OpenAI-compatible — change base_url and model name	Varies by provider and SDK	Full rewrite
Swahili text handling	Agglutinated forms plus expected English code-switching	Depends on provider tokeniser and prompt hygiene	You build normalisation, segmentation and evals
Token budget	Fixed tokeniser per model; measure with the Plugsky token calculator and chunk to fit	Varies by provider and model	You host and tune each tokeniser
Multilingual retrieval	plugsky-embed-multilingual available for cross-language RAG	Often needs a separate embedding vendor	You serve and maintain embeddings
Sovereignty	Cloud, VPC, on-prem and air-gapped with residency options	Usually US/EU public endpoints	You own the full stack

Frequently asked questions

Can Plugsky handle Swahili text?

Yes. The API accepts UTF-8 Swahili input on the OpenAI-compatible chat endpoint; output quality depends on the model, so compare two or three on your own prompts before choosing.

How do I estimate token usage for Swahili?

Swahili words carry prefixes for noun class, tense and person, so tokens often split at morpheme boundaries; code-switched English words tokenise like English, making mixed sentences uneven. Use the token calculator at /tools/llm-token-calculator before sizing context windows or chunk lengths.

Is English mixing acceptable in Swahili output?

Define the policy per product. Technical content commonly mixes English terms; consumer chat often does not. State the preference in the system prompt.

Which Swahili standard should I use?

Standard Kiswahili is the safest default. Keep Kenyan and Tanzanian vocabulary variants in separate eval slices if you serve both markets.

Is there a multilingual embedding model?

Yes — plugsky-embed-multilingual is part of the 30+ model catalogue and is built for cross-language retrieval. Keep one embedding model per vector collection.

Can I keep data in my region?

Plugsky supports cloud, VPC, on-prem and air-gapped deployment with data-residency options; confirm your requirements with the docs and the enterprise team.

How do I migrate an existing app?

Change base_url to https://api.plugsky.com/v1 and map the model name. Streaming, JSON mode, function calling and embeddings keep the same request shapes.

Is there a free plan?

Yes — the free plan includes two free models, plugsky-micro and plugsky-lite, with no card. A 14-day full-access trial unlocks the paid catalogue.