v0.3.3

What's Changed

• fix(tokenizer): encoding issues for special characters
• Remove trailing semicolon from README
• fix(client): bump default request timeout to 10 minutes

Full Changelog: v0.3.2...v0.3.3

v0.3.2

What's Changed

• upper bound pydantic version to 2.0 by @uripeled2 in #50
• update certifi to resolve https://github.com/anthropics/anthropic-sdk-python/security/dependabot/2

Full Changelog: v0.3.1...v0.3.2

v0.3.1

This release prevents the async tokenizer from crashing if called multiple times in the same process, #47.

v0.3.0 – ⚠️ BREAKING, fully rewritten library

In v0.3.0, we introduced a fully rewritten SDK.

The new version uses separate sync and async clients, unified streaming, typed params and structured response objects, and resource-oriented methods:

Sync before/after:

- client = anthropic.Client(os.environ["ANTHROPIC_API_KEY"])
+ client = anthropic.Anthropic(api_key=os.environ["ANTHROPIC_API_KEY"])
  # or, simply provide an ANTHROPIC_API_KEY environment variable:
+ client = anthropic.Anthropic();

- rsp = client.completion(**params)
- rsp["completion"]
+ rsp = client.completions.create(**params)
+ rsp.completion

Async before/after:

- client = anthropic.Client(os.environ["ANTHROPIC_API_KEY"])
+ client = anthropic.AsyncAnthropic(api_key=os.environ["ANTHROPIC_API_KEY"])

- await client.acompletion(**params)
+ await client.completions.create(**params)

The .completion_stream() and .acompletion_stream() methods have been removed;
simply pass stream=Trueto .completions.create().

Streaming responses are now incremental; the full text is not sent in each message,
as v0.3 sends the Anthropic-Version: 2023-06-01 header.

Example streaming diff

  import anthropic

- client = anthropic.Client(os.environ["ANTHROPIC_API_KEY"])
+ client = anthropic.Anthropic()

  # Streams are now incremental diffs of text
  # rather than sending the whole message every time:
  text = "
- stream = client.completion_stream(**params)
- for data in stream:
-     diff = data["completion"].replace(text, "")
-     text = data["completion"]
+ stream = client.completions.create(**params, stream=True)
+ for data in stream:
+     diff = data.completion # incremental text
+     text += data.completion
      print(diff, end="")

  print("Done. Final text is:")
  print(text)

Full Changelog: v0.2.10...v0.3.0

v0.2.10

What's Changed

• Add version header

Full Changelog: v0.2.9...v0.2.10

v0.2.9

What's Changed

• Remove length check from automatic SDK validations. by @jenan-anthropic in #30
• fix: api_url concat by @nero19960329 in #27

New Contributors

• @jenan-anthropic made their first contribution in #30
• @nero19960329 made their first contribution in #27

Full Changelog: v0.2.8...v0.2.9

v0.2.8

What's Changed

• Remove length check from automatic SDK validations #30

Full Changelog: v0.2.7...v0.2.8

v0.2.7

What's Changed

• Fix: UnicodeEncodeError by @Anthony-Sarkis in #19

New Contributors

• @Anthony-Sarkis made their first contribution in #19

Full Changelog: v0.2.6...v0.2.7

v0.2.6

What's Changed

• Better async support by @mikelambert in #17
• Fail open when the tokenizer fails to load by @mikelambert in #16

Full Changelog: v0.2.5...v0.2.6

v0.2.5

Fix for async API when handling errors