Models · Swiss AI Initiative

Apertus v1.5 70B, run in Switzerland by Infomaniak.

Apertus v1.5 70B is an open-weight model by the Swiss AI Initiative that reads text and images, released under the Apache-2.0 licence. On Evanesca, Infomaniak runs it in Switzerland — outside the EU, under an EU adequacy decision: your prompts are processed there by an independent Swiss company, and Evanesca keeps no copy of them.

Checked on 20 Oct 2026

Apertus v1.5 70B

by the Swiss AI Initiative · Apache-2.0

Switzerland (Infomaniak) Checks 4/6

Input / M: €1.2936
Output / M: €4.62
Quick question: €0.0022

Prices per million tokens, VAT included. Quick question: 300 tokens in, 400 out.

The model at a glance

Maker: Swiss AI Initiative
Licence: Apache-2.0
Size: 72B parameters
Context: 100k tokens as the operator announces it
Type: Text + images
Tool calling: Yes measured by a real call
Reasoning: —
Run by: Infomaniak Infomaniak, Geneva — an independent Swiss company, no US parent
Where it runs: Switzerland (Infomaniak)
Checks: 4/6 · Security, Benchmarks: not run yet Passed: Licence, Provenance, GDPR and Live test
Price: €1.2936 per million input tokens · €4.62 per million output tokens VAT included; reasoning tokens count as output
Open weights: Yes published by the maker: an operator can run the model
Model card: huggingface.co/swiss-ai/Apertus-v1.5-70B

Where your prompts go

Here is the route of a message sent to Apertus v1.5 70B, end to end. No call to the model leaves Europe: not as a backup, not when servers are busy.

From your browser to EvanescaEvanesca passes your message straight on to Infomaniak. It stores neither the prompt nor the answer: both stay in memory only while the answer is written.
Infomaniak, SwitzerlandInfomaniak runs Apertus v1.5 70B in Switzerland — outside the EU, but under an adequacy decision: the EU recognises its protection of personal data as adequate. It is an independent Swiss company with no US parent: out of the CLOUD Act's direct reach. Infomaniak states that queries are “neither recorded nor used to train models” (Infomaniak AI Tools).
Back to you, with its passportThe answer comes back with its passport: model, operator, country, and what was kept. Your history lives in your browser; about this message, Evanesca keeps only billing data — tokens and cost.

What it costs

€1.2936 per million input tokens and €4.62 per million output tokens, VAT included. Three typical exchanges on Apertus v1.5 70B:

COST OF AN EXCHANGE · VAT INCLUDED
Example	Tokens in	Tokens out	Cost
A quick question	300	400	€0.0022
A long document to summarise	15,000	800	€0.0231
A message late in a long conversation	40,000	2,000	€0.0610

€2 of credit covers about 890 quick questions on this model. Reasoning tokens count as output, so a model that thinks first can cost more than these examples. The exact cost of each answer shows under it. All prices.

Licence

The Swiss AI Initiative publishes Apertus v1.5 70B under the Apache-2.0 licence. Apache-2.0 is a permissive open-source licence: it allows commercial use.

The official model card is the reference.

Similar models

Same maker first, then the same type.

EU region of Mistral AI

Mistral Large 4

Mistral AI · Text + images · checks 4/6

€2.5133 · €7.7246 / M tokens

Germany (AKI.IO)

DeepSeek V4.1 Flash

DeepSeek · Text + images · checks 6/6

€1.008 · €3.024 / M tokens

Paris, France

Qwen3.8 27B

Qwen · Text + images · checks 6/6

€1.008 · €5.544 / M tokens

Paris, France

Qwen3.6 35B

Qwen · Text + images · checks 6/6

€0.42 · €2.52 / M tokens

Germany (AKI.IO)

Gemma 4 26B

Google · Text + images · checks 6/6

€0.168 · €0.84 / M tokens

EU region of Mistral AI

Mistral Small 4

Mistral AI · Text + images · checks 5/6

€0.2772 · €1.1088 / M tokens

Every model, with its operator and where it runs

Frequently asked questions

Where does Apertus v1.5 70B run on Evanesca?

In Switzerland, on Infomaniak's servers. Switzerland is outside the EU, but the European Commission recognises its protection of personal data as adequate (Decision 2000/518/EC): personal data can go there from the EU without additional guarantees. Infomaniak is an independent Swiss company with no US parent, so your prompts stay out of the CLOUD Act's direct reach. No call to the model leaves Europe: not as a backup, not when servers are busy.

Does Evanesca keep what I send to Apertus v1.5 70B?

No. The prompt and the answer stay in memory only while the answer is written, and your history lives in your browser; about your messages, Evanesca keeps only billing data — tokens and cost. Infomaniak states that queries are “neither recorded nor used to train models” (Infomaniak AI Tools).

How much does Apertus v1.5 70B cost?

€1.2936 per million input tokens and €4.62 per million output tokens, VAT included. A quick question — 300 tokens in, 400 out — costs about €0.0022. There is no subscription: credit starts at €2.

Can Apertus v1.5 70B call tools?

Yes: a real call through Infomaniak returned proper tool calls (function calling).

Which licence is Apertus v1.5 70B under?

Apache-2.0. Apache-2.0 is a permissive open-source licence: it allows commercial use.

Which checks has Apertus v1.5 70B passed?

4 of 6: Licence, Provenance, GDPR and Live test. Not run yet: Security and Benchmarks. Its passport shows the same stamps; how each check works.

Does the Swiss AI Initiative see what I type?

No. The Swiss AI Initiative published the weights, and Infomaniak runs them in Switzerland: your prompts go to Infomaniak's servers, not to the Swiss AI Initiative.