Lokalise
Lokalise · Ranked #6 of 7 in Translation APIs
Localization management platform with a translation REST API plus OTA SDKs for iOS, Android and Flutter; built for product teams, not per-call MT consumers.
TMS / localization platform
Overview
Lokalise is a translation and localization management platform (a TMS) built around an API-first workflow: source strings live as "keys" in projects, and a REST API plus webhooks let engineering teams automate uploading source content, downloading translations, and wiring localization into CI/CD. It targets product and engineering teams at scaling software companies who want to keep translation out of spreadsheets and inside their pipeline, with a strong emphasis on developer ergonomics (official SDKs in seven languages, CLI, OTA mobile SDKs, and deep integrations with GitHub, GitLab, Figma, and design/dev tooling). In late 2025 Lokalise repositioned heavily around AI, adding "AI orchestration" that routes content between multiple machine-translation engines and LLMs, plus a tiered Standard AI/MT and Pro AI model split.
As an API, Lokalise is mature and well-documented but deliberately throttled for a TMS rather than a high-throughput translation engine: every endpoint is capped at 6 requests/second per token and per IP, with a maximum of 10 concurrent requests per project and a 429 on overage. That is fine for sync jobs and CI automation but means Lokalise is not a drop-in real-time translation API the way Google Cloud Translation or DeepL are, it sits a layer above those engines, orchestrating them. Billing also shifted in the November 2025 plan revamp to be metered on "processed words" rather than stored keys, which is a recurring source of confusion and complaint.
Sentiment is strongly positive on usability and support (G2 ~4.7/5 across 700+ reviews, repeatedly ranked #1 in localization for support quality), but the critical thread is consistent and specific: pricing is seen as expensive and opaque, renewals have reportedly doubled, the word-based metering bills for manually edited words, and several users flag aggressive contract terms (90-day cancellation notice, auto-renewal). Glossary/terminology management is the most-cited functional weakness. Net: a polished, developer-friendly TMS that wins on UX, integrations, and support, but where buyers should scrutinize the commercial terms and not expect it to behave like a raw MT API.
How this score is derived
The APIbenchmarks Index is a weighted sum of four dimensions, each scored on an absolute 0–100 reference scale. See the methodology for every mapping.
| Dimension | Score | Weight | Contribution |
|---|---|---|---|
| Documentation & DXExtensive developer portal (developers.lokalise.com) with a full REST API v2 reference, rate-limit/exception-handling guidance, OpenAPI, and per-language SDK docs hosted on GitHub Pages. | 74 | 30% | 22.2 |
| ReliabilityPublic status page at status.lokalise.com tracks five components with historical incidents, and Lokalise advertises 99.9% uptime over the trailing 90 days, though no contractual SLA percentage is published outside Enterprise agreements. | 72 | 25% | 18.0 |
| Ecosystem & SDKsBroad ecosystem: official SDKs in seven languages, CLI, OTA mobile SDKs, and 50+ integrations spanning GitHub, GitLab, Figma, Sketch, Slack, and CMS/design tools. | 70 | 25% | 17.5 |
| AccessibilitySelf-serve signup with a 14-day free trial (no credit card) and transparent published plan prices, but full API access requires a paid plan and there is no perpetual free API tier. | 58 | 20% | 11.6 |
| APIbenchmarks Index (ABI) | 69.3 | ||
Table 1. Derivation of the ABI for Lokalise. Contribution = score × weight; the index is their sum.
At a glance
- Vendor
- Lokalise
- Pricing model
- Seat + plan tier
- Free tier
- No
- Official SDKs
- 11 languages
Pricing
| Explorer | $144/mo | ~5 advanced seats (max 10), 60,000 processed words/year, 240,000 Standard AI/MT words/year, 2 integrations, full API access (6 req/s cap). |
| Growth | $375/mo | ~10 advanced seats (max 15), 300,000 processed words/year, 50,000 Pro AI words/year, 3 integrations incl. Figma, more automation. |
| Advanced | $999/mo | 'Most popular.' ~15 advanced seats, 1,000,000 processed words/year, 150,000 Pro AI words/year, 5 integrations, 100 automations. |
| Enterprise | Custom | ~40+ advanced seats, 3,000,000 processed words/year, 400,000 Pro AI words/year, unlimited integrations, personalized AI models, bespoke security/support and SLA. |
Key features
- •REST API v2 with webhooks for two-way sync of keys and translations
- •AI orchestration across multiple MT engines and LLMs (Standard AI/MT and Pro AI tiers)
- •Over-the-air (OTA) updates for mobile apps via SDK
- •Translation memory and machine-translation pre-translation
- •In-context editing and Figma/Sketch design integrations
- •CI/CD and Git (GitHub/GitLab/Bitbucket) integrations
- •CLI tool (lokalise2) for scripted upload/download
- •QA checks for placeholders, HTML, and length
- •Branching and workflow/automation rules
- •Quality scoring and context handling in the Pro AI tier
Official SDKs
Strengths & trade-offs
- +Mature, well-documented REST API v2 with official SDKs in seven languages plus CLI and OTA mobile SDKs.
- +Deep integrations into dev and design workflows (GitHub, GitLab, Figma, Sketch, CI/CD), making it easy to embed in a pipeline.
- +Consistently praised, fast customer support, ranked #1 in localization for support across 700+ G2 reviews.
- +Intuitive UI/UX that non-technical translators and reviewers can use without much onboarding.
- +AI orchestration routes content across multiple MT engines and LLMs (Standard AI/MT plus Pro AI tiers) rather than locking you to one provider.
- +Transparent self-serve pricing with a 14-day no-credit-card trial.
- –Pricing is widely described as expensive and opaque; some users report renewals more than doubling.
- –November 2025 shift to word-based ('processed words') metering bills for every word edited, including manual edits, despite unlimited keys.
- –Aggressive commercial terms reported: 90-day cancellation notice and auto-renewal clauses that have caught customers off guard.
- –Glossary and terminology management is limited, no easy way to verify correct terminology was used in a given key.
- –API is hard-capped at 6 requests/second and 10 concurrent requests per project, so it is not suited to real-time/high-throughput translation.
- –Downgrading a plan can silently strip features (e.g. glossary) users didn't expect to lose.
What developers say
G2 4.7/5 · 700+ reviews
Users love the intuitive UI, integrations, and standout customer support, but a consistent minority criticize expensive, opaque word-based pricing and aggressive contract terms.
“Users consistently praise the intuitive interface and fast customer support; the UI is straightforward and easy to understand, making everything more efficient.”
Key figures
| API rate limit | 6 requests/second (per token and per IP) | Lokalise developer docs ↗ |
| Max concurrent requests per project | 10 (overage returns HTTP 429) | Lokalise developer docs ↗ |
| Advertised uptime (trailing 90 days) | 99.9% | Lokalise status / marketing ↗ |
| Explorer plan price | $144/month | Lokalise pricing page ↗ |
| Advanced plan price | $999/month | Lokalise pricing page ↗ |
| Aggregate user rating | 4.7/5 across 700+ reviews | G2 ↗ |
Compare Lokalise head to head
Sources
- https://lokalise.com/pricing/
- https://docs.lokalise.com/en/articles/11694835-new-price-plans-everything-you-should-know
- https://developers.lokalise.com/reference/api-rate-limits
- https://status.lokalise.com/uptime
- https://www.g2.com/products/lokalise/reviews
- https://www.capterra.com/p/162858/Lokalise/reviews/
- https://www.trustpilot.com/review/lokalise.com
- https://github.com/lokalise/node-lokalise-api
- https://lokalise.com/product/api-translation/
Figures last verified 2026-06-27. Spotted an error? corrections@apibenchmarks.com