Skip to main content

Choosing the Right Model

The right model depends on your task, compliance needs, and budget. All Schatzi AI models run exclusively on Swiss infrastructure, so your data never leaves Switzerland.

Quick Decision Guide

"I just want to draft an email or a message"

→ Apertus Swiss LLM - Large

"I need Swiss/EU compliance and transparency"

→ Apertus Swiss LLM - Large

"I need complex reasoning and agent workflows"

→ Chat & Document Analysis & Reasoning - Large

"I need up-to-date web search and current information"

→ Search, Chat & Analysis - Small

"I need to analyse documents and images"

→ Chat & Document Analysis & Reasoning - Large

"I need maximum capability across everything"

→ Chat & Document Analysis & Reasoning - Large

Model Capabilities at a Glance

API-only models are accessible via the REST API but are not visible in the Chat UI.

ModelCategoryVisionReasoningWeb SearchFunction CallingContext WindowAvailability
Apertus Swiss LLM - LargeSwiss LLM (AI Act Compliant)NoNoNoNo65,536Chat UI & API
Chat & Document Analysis & Reasoning - LargeReasoning & Problem-SolvingYesYesNoYesNot specifiedAPI only
Document Analysis - SmallVision & Document AnalysisYesNoNoYes32,768API only
Document Analysis - Xtra SmallVision & Document AnalysisYesNoNoNo16,384API only
Fast Reasoning & Instruction Following - SmallReasoning & Problem-SolvingNoNoNoYes32,768API only
Reasoning & Problem Solving - SmallReasoning & Problem-SolvingNoYesNoYes32,768API only
Llama 4 Maverick multi modal - SmallVision & Document AnalysisYesNoNoYes32,768API only
Reasoning & Agent tasks - LargeReasoning & Problem-SolvingNoYesNoYes65,536API only
Reasoning & Problem Solving - MediumReasoning & Problem-SolvingNoYesNoYes32,768API only
Reasoning & Problem Solving - Xtra LargeReasoning & Problem-SolvingNoNoNoNoNot specifiedAPI only
Reasoning & Tool Use - Large (GLM-4.5 Air)Reasoning & Problem-SolvingNoYesNoYes131,072API only
Search, Chat & Analysis - SmallChat & General PurposeYesNoYesNoNot specifiedAPI only
Chat, Document Analysis & Agent tasks - Xtra LargeReasoning & Problem-SolvingYesYesYesYes250,000Chat UI & API
Document Analysis & OCR - Small (DeepSeek OCR)Vision & Document AnalysisYesNoNoNo8,192API only
Chat, Multi-lingual, Coding & function calling - SmallMultilingualNoNoNoYes128,000Chat UI & API
Chat, Document Analysis, Coding & Reasoning - Xtra LargeReasoning & Problem-SolvingYesYesNoYes1,000,000Chat UI & API
Chat, Vision, Document Analysis & Reasoning - MediumReasoning & Problem-SolvingYesYesNoYes256,000Chat UI & API
inference-miner-u25Vision & Document AnalysisNoNoNoNoNot specifiedAPI only
Reasoning & Tool Use - Xtra Large (GLM-5.2)Reasoning & Problem-SolvingNoYesNoYesNot specifiedChat UI & API
Apertus Swiss LLM - Large (v1.5)Swiss LLM (AI Act Compliant)NoNoNoNo65,536API only

Pricing Overview

Loading prices...

Pricing Notice

Pricing is subject to change at our discretion.

For how tokens are calculated and billed, see Understanding Tokens.

General Tips

Start Simple

Begin with a cost-effective model such as Apertus Swiss LLM - Large for routine tasks, and move to a more capable model when you need vision or reasoning. See the Lite Plan for entry-level access.

Match Compliance Requirements

For strict data sovereignty and AI Act compliance, Apertus Swiss LLM - Large is the recommended choice. Learn more about our Data Protection standards.

Consider Context Window

Match the model's context window to your document length — longer documents need a larger context window.

When to Upgrade / Downgrade

  • Upgrade to Chat & Document Analysis & Reasoning - Large or Chat & Document Analysis & Reasoning - Large when a task needs deeper reasoning, agent workflows, or multi-modal capability.
  • Downgrade to Apertus Swiss LLM - Large for routine chat and drafting to optimise for speed and cost.
Pro Tip

Most users find 2–3 models cover 90% of their needs — typically Apertus Swiss LLM - Large for quick tasks, Chat & Document Analysis & Reasoning - Large for specialised work, and Apertus Swiss LLM - Large for compliance-sensitive operations. Monitor usage in the dashboard to optimise over time.