Swedish-hosted open models, competitively priced
This is the price list for the Grunden API. Prices are in EUR, excluding VAT. Language models are priced per one million input and output tokens. Every model on this page runs in Sweden and can be used in production.
Language models
| Model | Capabilities | Context | Accepts | Returns | Reasoning | Input /1M | Cached input /1M | Output /1M |
|---|---|---|---|---|---|---|---|---|
deepseek-v4.1-flash-1017 |
chatreasoningtoolsvision | 1024k | text, image | text | low · high · max | €0.50 | €0.10 | €1.50 |
These models bill cached input at a lower rate than regular input.
Async delivery is billed at half these prices. The result arrives by webhook within 12 hours.
Other services
| Model | Service | Endpoint | Price |
|---|---|---|---|
bge-m3 |
Embeddings | /v1/embeddings |
€0.01 per 1M input tokens |
bge-reranker-v2-m3 |
Reranking | /v1/rerank |
€0.05 per 1,000 queries |
gliner-multi-pii |
Personal data detection | /v1/pii/redact |
€0.02 per 1M input characters |
kb-whisper-large |
Speech to text | /v1/audio/transcriptions |
€0.50 per audio hour |
pianissimo-da |
Speech to text | /v1/audio/transcriptions |
€0.25 per audio hour |
pianissimo-sv |
Speech to text | /v1/audio/transcriptions |
€0.25 per audio hour |
chatterbox |
Text to speech | /v1/audio/speech |
€90.00 per 1M input characters |
Last updated 2026-11-21 21:40