Explore supported chat, vision, image generation, embeddings, speech, transcription and reranking models. APIs, supplier availability and routing options vary by model.
POST /v1/chat/completions
{
"model": "auto",
"region": "EU",
"messages": [{
"role": "user",
"content": "Hello!"
}]
}EU-only routing is available for supported models; review execution and data-handling boundaries.
Unified access to every leading model family across multiple providers.
| Family | Modalities | Max context | Suppliers | Geo-filter |
|---|---|---|---|---|
| Chat & Completion Alibaba · Anthropic · CloudSigma +8 |
Chat & Completion | 2M | 11 | |
| Vision & Multimodal Alibaba · Mistral AI |
Vision & Multimodal | 131K | 2 | |
| OCR & Documents CloudSigma · Mistral AI |
OCR & Documents | 33K | 2 | |
| Document Conversion CloudSigma |
Document Conversion | - | 1 | |
| Image Generation OpenAI |
Image Generation | - | 1 | |
| Embedding CloudSigma |
Embedding | - | 1 | |
| Reranking CloudSigma |
Reranking | - | 1 | |
| Text-to-Speech CloudSigma |
Text-to-Speech | - | 1 | |
| Transcription OpenAI |
Transcription | - | 1 | |
| Realtime Transcription OpenAI |
Realtime Audio | 33K | 1 | |
| Speaker Recognition CloudSigma |
Speaker Recognition | - | 1 | |
| Audio Understanding CloudSigma |
Audio Understanding | - | 1 |
Rates shown are evaluated now in each schedule's timezone. Open the pricing page for the complete daily and weekday schedule.
Apply supported access and routing controls to model requests.
Where compatible alternatives are configured, eligible failures can retry on another supplier before meaningful output. Applications must still handle terminal errors.
Use an EU-only key policy for supported model lanes. Review inference, logging, storage and fallback boundaries separately.
Prefer specific providers for cost or compliance reasons? Restrict keys to named suppliers while keeping automatic failover within your approved list.
Use supported chat, vision, image, embedding, speech, transcription and reranking capabilities through their documented endpoints. Compatibility and billing units vary.
Use documented OpenAI-compatible chat, Anthropic-compatible Messages and native specialist interfaces. Check the supported endpoint and model matrix when migrating.
Review model and key usage and cost breakdowns. Set spending controls and discuss hard-budget enforcement scope during enterprise evaluation.
Configure an EU-only key policy and choose an eligible model. Routing uses the active provider geography catalogue and rejects requests where no permitted route is available.
Review permitted execution locations alongside provider logging, storage and fallback policies. Route eligibility is not a blanket compliance or end-to-end residency guarantee.
Explore access controls# EU-only chat request
curl https://taas.cloudsigma.com/v1/chat/completions \
-H "Authorization: Bearer $TAAS_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "auto",
"region": "EU",
"messages": [
{"role": "user", "content": "Summarise this contract."}
]
}'
# Inspect documented response metadata and request logs.Routing can use health signals and compatible alternative suppliers. Retry eligibility depends on the endpoint, error, policy and whether output has already started.
Failover respects your geo and supplier restrictions: if your key is locked to EU suppliers, failover only considers other EU-approved routes.
Read the developer guide# Python — documented OpenAI-compatible chat
from openai import OpenAI
client = OpenAI(
api_key="your-taas-key",
base_url="https://taas.cloudsigma.com/v1"
)
# Handle errors according to the endpoint documentation
response = client.chat.completions.create(
model="auto",
messages=[{"role": "user",
"content": "Draft a privacy notice."}]
)
print(response.choices[0].message.content)Explore the public catalogue, review endpoint compatibility and establish access, spending and location requirements.