DeepL pricing

About

DeepL is a German AI company that turned a reputation for unusually fluent machine translation into a multi-product business. It started as Linguee — a bilingual dictionary launched in 2009 — and shipped the DeepL Translator in 2017. Today it sells four things off one engine: the DeepL Translator apps, DeepL Write (an AI writing and grammar tool), DeepL Voice (speech translation), and the DeepL API for developers.

The company is privately held, headquartered in Cologne, and raised in May 2024 at a roughly 2B valuation in a round led by Index Ventures — putting it among Europe’s most valuable AI companies. Crucially for pricing, DeepL serves two very different buyers with two very different meters: knowledge workers who buy seats in the apps, and developers who buy characters through the API. Understanding DeepL’s pricing means understanding that split.

For the most current information on DeepL’s pricing, visit DeepL Pro and the DeepL API page.

Pricing summary : How DeepL’s pricing model works

DeepL runs a hybrid model that is really two pricing systems sharing one translation engine.

1. Per-seat Pro subscriptions (the apps). Knowledge workers buy a seat. The ladder is Free, then Individual at $8.74/user/mo, Team at $28.74/user/mo, and Business at $57.49/user/mo — all on annual billing, with monthly billing about 16% higher. Each tier raises three limits at once: characters per month, the number of file/document translations, and glossary capacity. Enterprise is sales-quoted.

2. Per-character API (developers). The API is pure usage-based: you pay per source character translated. There is a free tier for testing and a paid tier that adds a small fixed monthly base fee on top of per-character usage. The two figures you will see quoted are the long-running API Pro model — $5.49/month base + $25.00 per 1,000,000 characters, with 500,000 free characters/month — and DeepL’s newer Growth packaging at $26/month (annual) for ~1,000,000 characters/month with overage at $27.50 per 1,000,000 characters.

What makes this different: most AI vendors pick one meter and stick with it. DeepL deliberately keeps two — seats for humans, characters for machines — because the same translation is worth a flat, predictable seat fee to a marketing team and a metered per-character fee to an app developer. The engine is identical; only the billing unit changes with the buyer.

Pricing by product

Sales motions across products: self-serve and PLG on the apps and the API (sign up, pay by card, no call), with a sales-led Enterprise motion for SSO, custom SLAs, and BYOK. The API and the lower app tiers are fully self-serve.

Hidden costs : What DeepL users actually pay

The headline rates hide three real cost drivers.

The API base fee. The per-character rate looks cheap, but the paid API tier charges a fixed monthly fee — $5.49/mo on the classic API Pro model, $26/mo on Growth — before any usage. At low volume that fixed fee dominates: translate 500,000 characters in a month and the base fee can be roughly 30% of your total bill. The API is cheap per character but not cheap to start.

Seat sprawl on the apps. Subscriptions are per user. A 20-person team on Team pays 20 × $28.74/user/mo whether or not everyone translates daily. There is no pooled-character option below Business, so light users still cost a full seat.

Tier-jumping for documents and glossaries. It is rarely the character limit that forces an upgrade — it is the document count (3 → 20 → 100 files/mo) and glossary capacity. A small team that translates many PDFs can be pushed to Business by file limits alone, not volume.

Want to estimate your own DeepL bill? Use the DeepL pricing calculator to model API characters and seat counts side by side.

Pricing evolution : DeepL pricing history and changes

Cadence

Tracked range: 2017–present. The per-character API meter has been the constant since 2018; what has moved is how the fixed component is packaged (flat base fee → bundled annual allowance).

Notable changes

• 2017 — DeepL Translator launches with a Free tier and paid DeepL Pro seats.
• 2018 — DeepL API opens with per-character metering and a free allowance.
• 2024 — DeepL raises at a roughly 2B valuation and expands into DeepL Write and DeepL Voice, all billable through one account.
• 2026 — The API’s fixed component is repackaged as Growth ($26/mo for ~1M characters/mo + $27.50 per 1,000,000 overage) alongside the classic API Pro framing ($5.49/mo + $25.00 per 1,000,000).

What’s unique : DeepL’s distinctive pricing mechanics

1. Two meters, one engine. DeepL bills the exact same translation model two completely different ways — per seat for app users, per character for API developers. Few vendors run a clean dual meter this deliberately; it lets DeepL price for predictability (seats) and for scale (characters) without forcing either buyer into the wrong model.

2. A fixed base fee on a usage product. The API is usage-based but not pay-as-you-go in the purest sense: there is always a monthly floor. That floor makes revenue predictable for DeepL and filters out trivial usage, but it makes DeepL expensive per character at the low end — a deliberate trade.

3. Limits stacked three-deep. App tiers don’t gate on one number. Characters, document count, and glossary size all move together, so the “right” tier depends on your workflow shape, not just your volume. A heavy-document, low-character user and a high-character, no-document user can land on different tiers at the same spend.

Strengths & weaknesses

Billing UX : DeepL billing controls and transparency

• Billing controls — Self-serve sign-up and card payment for the apps and the API; annual vs monthly billing toggle (annual saves ~16%). Enterprise moves to invoicing and contracts.
• Usage visibility — The API account dashboard shows character consumption against your allowance, and DeepL exposes a usage/quota endpoint so developers can monitor spend programmatically before hitting overage.
• Payment options — Credit/debit card for self-serve tiers; invoicing, POs, and custom terms on Enterprise. Per-seat plans bill per active user with central billing on Team and above.

Strategic wins : Why DeepL’s pricing decisions worked

1. Matching the meter to the buyer

By selling seats to humans and characters to machines, DeepL avoided the classic mistake of forcing developers onto per-seat pricing or knowledge workers onto unpredictable metered bills. Each buyer gets the model that fits their mental cost model. See choosing the right usage metric for why the unit matters more than the rate.

2. A free tier that actually converts

Both free tiers are usable, not crippled trials — 50,000 characters/mo in the app, a real free API allowance. That turns DeepL into the default first click for translation, and the upgrade triggers (characters, documents, glossary) are natural growth walls. Related: how AI companies structure pricing.

3. A base fee that stabilizes usage revenue

The fixed monthly API fee gives DeepL a predictable revenue floor on top of variable usage — the same hybrid logic many infrastructure vendors have adopted. It trades some low-end developer goodwill for revenue durability. See outcome- and usage-based pricing trends and the introduction to usage-based pricing for the underlying model.

Areas to improve : Gaps in DeepL’s pricing approach

1. The base fee surprises small developers

The fixed API fee isn’t always obvious until the first invoice, and at low volume it dwarfs usage. Surfacing an honest “your effective per-character cost at this volume” estimate would reduce bill shock.

2. No pooled characters for small teams

Below Business, every seat is a separate character bucket. A small team with a few heavy translators and many occasional ones overpays. A pooled-allowance option would fit real team usage better.

3. Two pricing pages, one mental model gap

Buyers who need both the apps and the API have to reconcile two pricing systems and two billing units. A unified view — “here is your blended cost across seats and characters” — would help mixed buyers reason about total spend.

Key takeaways

1. Two meters, one engine. DeepL bills the same translation model per seat (apps) and per character (API) — pick the side that matches your buyer.
2. The API floor is the catch. A fixed monthly base fee sits under per-character usage, so the API is cheap at scale but expensive to start.
3. Tiers gate on workflow, not just volume. Document counts and glossary size move app tiers as much as characters do.
4. Free tiers are real conversion engines. Usable free allowances on both sides make DeepL the default first click for translation.
5. Transparency is a competitive asset. Public, self-serve pricing on both apps and API lowers the barrier in a category where rivals often hide rates.

UBP implications

1. A base fee can rescue a usage product’s economics — DeepL’s fixed API floor turns volatile per-character revenue into a predictable stream, at the cost of low-end appeal.
2. Segment the meter, not just the price — selling characters to developers and seats to teams shows that the right unit can matter more than the rate when buyers differ.
3. Stacked limits shape upgrades — gating on characters, documents, and glossary together gives more upgrade triggers than a single dimension, but raises the risk of upgrades that feel unrelated to value.

Sources

• DeepL Pro subscription pricing (accessed 2026-06-16)
• DeepL API pricing (accessed 2026-06-16)
• DeepL Help Center — usage count and billing in the API (accessed 2026-06-16)
• eesel AI — DeepL pricing in 2026 (accessed 2026-06-16)
• CheckThat.ai — DeepL pricing 2026 (accessed 2026-06-16)
• Browse the full pricing blueprint for comparable companies.

Bottom line

DeepL is a rare AI vendor that bills one engine two ways: predictable per-seat Pro subscriptions for the apps (from $8.74/user/mo) and a usage-based, per-character API for developers. The model is unusually transparent, with real free tiers on both sides — but the API’s fixed monthly base fee makes it expensive for low-volume developers, and app tiers gate on document and glossary limits as much as on characters. Match the meter to your buyer and the economics are clean.

Want to compare DeepL against other AI translation and language companies? Browse the pricing blueprint.

Pricing timeline : Major events on a vertical axis

Each milestone below corresponds to a public pricing change, product launch, or material adjustment. Major events use a filled marker; minor adjustments use a faded one.

Current ladder: per-seat Pro + per-character API

Jun 2026

Pro subscriptions ladder Free / Individual $8.74 / Team $28.74 / Business $57.49 per user/mo (annual); the API offers a free tier plus a paid tier (base fee + per-character usage), with Growth packaging at $26/mo + $27.50 per 1M characters.

$2B valuation raise; product line expands

May 2024

DeepL raised at a roughly 2B valuation (Index Ventures lead) and broadened beyond translation into DeepL Write and, later, DeepL Voice — all billable through the same account.

DeepL API introduced with per-character metering

Mar 2018

DeepL opened a developer API priced by usage — a free character allowance plus a paid tier metered per source character, the meter that still defines API pricing today.

DeepL launches translator with a free + Pro split

Aug 2017

DeepL launched out of Linguee, offering a free web translator and paid DeepL Pro subscriptions for higher limits, document translation, and data privacy.