anthropic 1.12.0 on Python PyPI

1.12.0 (2026-10-07)

api: add claude-haiku-5-5 and typed computer and browser toolset tool calls (f6346e6)
api: add disabled to thinking types in model capabilities (78d815d)
api: add display_name to RBAC roles and deprecate name (f63f99c)
api: add include_default parameter to list workspaces (5d4b37b)
api: add lifecycle stage fields and filter to /v1/models (0c96c02)
api: add line to model objects (a85c649)
api: add url_sources to the Managed Agents web_fetch tool config (78d815d)
api: add web search and code execution support to model capabilities (78d815d)

api: send the beta header by default when listing spend limits (e2690b4)
client: send empty strings in query params and multipart forms (80e69f1)
tools: keep a local edit made after a memory upload whose response was lost (#990) (78d815d)

client: track whether the client timeout was set explicitly (36a223f)
docs: correct example groups in rate limit list description (71fcf2f)
docs: describe a federation rule's target by its type (78d815d)
docs: fix example IDs in sessions, agents and vault credentials (78d815d)
docs: update Managed Agents multiagent and thread descriptions (78d815d)
docs: update the activity summaries endpoint description (d9bd4bf)
docs: update the description of the Model line field (4444463)
internal: add REVIEW.md with review instructions (78d815d)
internal: format the code blocks of the docs with one run of ruff (78d815d)
internal: refactor deprecated models (7b0c7d5)
tests: run tests skipped for a query param bug that is now fixed (d606929)
tests: run tests that an older mock server could not serve (bffd222)
tests: run the batch results raw and streaming response tests (#978) (78d815d)
tests: run the tests skipped for incomplete example responses (06a75bc)