Comparison

Frontière vs Together AI: the jurisdiction question

Together AI and Frontière are both curated, but they sit under different legal systems, and on price Together AI wins. Together Computer, Inc. is a San Francisco company running its own infrastructure, which gives it consistent performance and a broad catalog including fine-tuning and training — on Llama 3.3 70B it publishes €0.90 ($1.04) per million input tokens against our €1.26. What owning your own hardware does not change is who can compel access to it: a US-incorporated company remains subject to US legal process under the CLOUD Act. Frontière is a French company routing to EU providers with no non-EU capital control, labeling every model's jurisdiction before you call it, and storing no prompts. Pick Together AI for breadth, training and price; pick Frontière when you have to answer for jurisdiction rather than just for data location.

Side by side

Every row below comes from the other side's own published pages, checked on 2026-08-06 and linked at the end of this section — not from a third-party comparison.

CriterionFrontièreTogether AI
Operating companyFLEECE AI SASUTogether Computer, Inc.
HeadquartersFranceSan Francisco, California, United States
Governing lawFrench / EU lawState of California
US CLOUD Act exposureNone on the 12 sovereign models (EU companies, no non-EU control)Yes — US-incorporated company, regardless of where inference runs
How the catalog worksCurated catalog, 13 models live, each pinned to a named providerDirect provider running its own infrastructure, plus fine-tuning and training
Llama 3.3 70B, per 1M tokens (in / out)€1.26 / €1.26€0.90 ($1.04) / €0.90 ($1.04)
Prompt storageNever stored — only token counts and the amount chargedZero Data Retention available as an opt-in setting; no training on customer data without explicit consent
Jurisdiction visible per modelYes — sovereign / fast access, on the card and in the API responseNo EU data residency documented in the privacy policy; transfers covered by standard contractual clauses

Prices published in USD, shown converted at 1 USD = 0.8655 EUR (rate of 2026-08-05) alongside the original figure. Frontière publishes in EUR, margin included — the amount actually debited.

Sources, checked on 2026-08-06: together.ai — pricing · together.ai — privacy policy · together.ai — terms of service

Where Together AI is the better choice

  • It runs its own infrastructure, so performance is consistent and does not depend on which third party picked up your request.
  • Cheaper on Llama 3.3 70B, and its catalog is broader than ours across model families and sizes.
  • Fine-tuning and training are first-class products there; we only serve inference on a curated catalog.
  • Zero Data Retention is available as an explicit setting, and the privacy policy states customer data is not used for training without opt-in consent.

Where Frontière is the better choice

  • The operating company is French. There is no US parent, licensee or affiliate through which a US order could reach your data on the sovereign tier.
  • Every model carries its jurisdiction before you call it, in the catalog and in the API's own response.
  • API prompts and completions are never written to our database — only the token counts billing requires, and that is the default rather than a setting to find. Our web console can save a conversation, but only when you explicitly sync it.
  • Prices are published in euros, margin included, so the figure you compare is the figure you are billed.

How to choose, concretely

The two are not mutually exclusive, and for most teams the honest answer is to use both:

  • Fine-tuning, custom weights, high-volume commodity inference: Together AI does things we deliberately do not.
  • Regulated, personal or client data — anything a DPO signs off on: pick the tier where corporate jurisdiction is settled, not just the contractual safeguards.
  • Zero Data Retention and EU jurisdiction answer different questions. The first limits what a provider keeps; the second limits who can compel it. A workload that needs both needs the second one to be structural.

GDPR vs. the CLOUD Act →

FAQ

Is Together AI GDPR-compliant?

Together AI documents the mechanisms GDPR asks for: standard contractual clauses for international transfers, data subject rights, an opt-in Zero Data Retention setting, and no training on customer data without consent. What none of that addresses is jurisdictional reach: Together Computer, Inc. is a California company, so US legal process under the CLOUD Act applies to data it controls. Compliance paperwork and jurisdictional exposure are separate axes.

Is Frontière cheaper than Together AI?

Not on Llama 3.3 70B: Together AI publishes €0.90 ($1.04) per million input tokens against €1.26 on Frontière. Our prices include a fixed margin over what named EU providers charge us, and sovereign EU capacity is not the cheapest capacity on the market. The argument for Frontière is jurisdiction, not price.

Together AI offers Zero Data Retention — isn't that the same thing?

It answers a different question. Zero Data Retention limits what the provider keeps of your prompts, which is genuinely useful. Jurisdiction determines who can compel a company to hand over what it does hold or process, and under which legal process. A US company with retention disabled is still a US company. On Frontière, API prompts are not stored either — and the operating company is not reachable by a US court order in the first place. The two properties are complementary, not interchangeable.

Does Together AI offer EU data residency?

Its published privacy policy does not document an EU residency or EU-region inference option, and states data may be transferred and maintained outside your country under standard contractual clauses. If EU-region processing matters to your policy, that is worth confirming with them directly rather than inferring it — we would rather point you at the question than answer it on their behalf.

See what we actually serve

Thirteen models live, each with its publisher, licence, hosting jurisdiction and real price.