Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -1549,7 +1549,7 @@ lenz extract pitch_deck.pdf --focus "market size"
## Multi-language output

The Lenz API returns prose fields (atomic claim, executive summary, debate, panel
reasoning) in any of 13 languages. Pass `language=` on `verify`, `verify_and_wait`,
reasoning) in any of 39 languages. Pass `language=` on `verify`, `verify_and_wait`,
`verify_batch`, `assess`, `extract`, or `ask.send`. Verdict labels stay English
regardless of language. Serbian (`sr`) is answered in the Latin script. On `extract`, `language` and `focus` are independent —
a focus written in any language selects claims emitted in `language`.
Expand Down
2 changes: 1 addition & 1 deletion openapi.json
Original file line number Diff line number Diff line change
Expand Up @@ -350,7 +350,7 @@
"x-codeSamples": [
{
"lang": "Python",
"source": "from lenz_io import Lenz\n\nclient = Lenz(api_key=\"lenz_...\")\nv = client.verify_and_wait(claim=\"Sharks don't get cancer\")\n# Add language=\"es\" for Spanish output (12 languages supported).\n# Add depth=\"low\" for a shallower check that returns sooner.\nprint(v.verdict, v.lenz_score)\n# False 2\n"
"source": "from lenz_io import Lenz\n\nclient = Lenz(api_key=\"lenz_...\")\nv = client.verify_and_wait(claim=\"Sharks don't get cancer\")\n# Add language=\"es\" for Spanish output (39 languages supported).\n# Add depth=\"low\" for a shallower check that returns sooner.\nprint(v.verdict, v.lenz_score)\n# False 2\n"
},
{
"lang": "TypeScript",
Expand Down
2 changes: 1 addition & 1 deletion src/lenz_io/client.py
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@
* Request methods (verify, assess, extract, ask, …) take ``language=''``
as their default. Sending an empty string means "do NOT include the
field in the request body" — preserves byte-identical behavior for
existing English callers. Set ``language='es'`` (or any of the 12
existing English callers. Set ``language='es'`` (or any of the 39
supported codes) to receive prose fields in that language. ``assess``,
``verify`` (and ``verify_and_wait``), ``review`` (and ``review_and_wait``),
``extract`` and ``ask.send`` also take ``language='auto'``: the answer comes
Expand Down
2 changes: 1 addition & 1 deletion tests/test_language.py
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@
* ``language=''`` (default) MUST omit the field from the request body.
This is the CRITICAL regression invariant: every existing English
caller's wire format must stay byte-identical after this change.
* ``language='es' | 'de' | …`` (any of 12 supported codes) MUST send
* ``language='es' | 'de' | …`` (any of 39 supported codes) MUST send
``"language": "<code>"`` in the JSON body.
* ``language='auto'`` (``assess``, ``verify``, ``ask.send``) is sent as
``"language": "auto"``, unchanged. The SDK does no validation of its own.
Expand Down
Loading