01 / Quickstart

Your first request

No account or key needed until Dec 1, 2026. Copy the request below and send it straight to the chat endpoint.

terminal

curl https://api.unboringspot.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3:8b",
    "messages": [
      { "role": "user", "content": "Explain edge inference in one sentence." }
    ],
    "think": false,
    "stream": false,
    "options": {
      "temperature": 0,
      "num_predict": 50
    }
  }'

Not OpenAI-compatible. Call these routes directly. OpenAI SDKs and base-URL swaps are not supported.

02 / Authentication

No authentication required

The API is free and open for production until Dec 1, 2026. Call any route directly — no Authorization header, no account, no key.

From Dec 1, 2026, requests will need a Bearer token from an account. Watch the changelog for the exact rollout date.

03 / Text

Chat and completion

Send a messages array to the chat route. Set think: false to skip Qwen3's reasoning step and get a direct answer back faster. Set stream: true instead of false to receive server-sent events as the response is generated.

model: qwen3:8bPOST /v1/chat/completionsJSON

request.sh

curl https://api.unboringspot.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3:8b",
    "messages": [
      { "role": "system", "content": "You are a concise assistant." },
      { "role": "user", "content": "Summarize this support thread." }
    ],
    "think": false,
    "stream": false,
    "options": {
      "temperature": 0,
      "num_predict": 50
    }
  }'

response.json

{
  "model": "qwen3:8b",
  "message": {
    "role": "assistant",
    "content": "Edge inference keeps latency low."
  },
  "done": true
}

04 / Vision

Send an image with a prompt

model: qwen-vlmcoming soon

Qwen-VLM is not live yet. Vision requests on the chat route will return an error until this model ships — watch the changelog for the rollout.

05 / Images

Generate an image

model: flux-schnellcoming soon

Flux Schnell is not live yet. Image generation is not available on the API until this model ships.

06 / Audio

Transcribe audio

model: whispercoming soon

Whisper is not live yet. Audio transcription is not available on the API until this model ships.

07 / Models

Model reference

Model strings are stable. Switch models by changing the model field in the request body.

Model	Modality	Route
`qwen3:8b` Qwen3 8B	Text	POST /v1/chat/completions
`qwen-vlm` Qwen-VLM	Vision	Coming soon
`flux-schnell` Flux Schnell	Image	Coming soon
`whisper` Whisper	Audio	Coming soon

All models are free to use while the API is open, until Dec 1, 2026. See the FAQ for what happens after.

08 / Errors

Error responses

Errors return a JSON body with a code and message. Use the table below to decide whether to retry.

Status	Meaning	Action
400	Bad request	Malformed JSON, a missing model, or an invalid field.
401	Unauthorized	The API key is missing or invalid.
404	Unknown model	The model string is not in the public lineup.
413	File too large	The audio upload is larger than the route allows.
429	Rate limited	Too many requests. Wait for the reset window and retry.
500 / 503	Service error	Retry with backoff. Email us if the error persists.

09 / Rate limits

Requests and concurrency

requests per minute per IP during the free, open period. Media requests count as one request.

Higher

limits are available for production workloads. Tell us what you are building.

Need more throughput or a model that is not listed? hello@unboringspot.com.

Start building

Make your first request.

Pick a model and call the route. No account, no key, no SDK, no sales call — free and open until Dec 1, 2026.

Try the playground Read the FAQ