What is Hyperlocalise? - hyperlocalise

hyperlocalise helps you generate local translation drafts, optionally sync with your TMS, and track what still needs review.

Platform focus

LLM provider layer: OpenAI, Azure OpenAI, Gemini, Anthropic, AWS Bedrock, LM Studio, Groq, Ollama
TMS adapters (experimental): Crowdin, LILT AI, Lokalise, Phrase, POEditor, Smartling
Eval framework (experimental): quality + regression checks across locales/models
CI-ready status labels (experimental): ready / needs review / missing
Plan-first + lockfile: deterministic runs and reviewable diffs

Feature graph

Who this is for

Use this CLI if you:

keep translation files in your repository,
want AI-generated drafts as a starting point,
want to choose between zero-human workflows and optional human curation in your TMS.

Core workflow

Stage	Action	Why it matters
1	`init`	Scaffold `i18n.yml` and bootstrap defaults.
2	Configure `i18n config`	Define locales, buckets, and LLM/storage settings.
3	`run --dry-run`	Validate plan and detect issues before writing drafts.
4	`run`	Generate local draft translations.
5	Release from local repo	Zero-human path when your process allows direct publish from generated outputs.
6 (optional)	`sync push` (experimental)	Upload local changes to your TMS for curation workflows.
7 (optional)	Curate in TMS	Human review and correction in your translation platform.
8 (optional)	`sync pull` (experimental)	Bring curated translations back into the repository.
9	`status`	Measure completion and unresolved work in either workflow path.

Start in 10 minutes

Common next steps

Learn command behavior in commands overview.
Configure provider credentials in provider credentials.
Understand sync behavior in storage overview.
Review feature maturity in the stability matrix.

Why Hyperlocalise for AI Localization Workflows