Skip to content

Releases: hmarr/openai-chat-tokens

v0.2.8

28 Sep 11:21
afe0e5b
Compare
Choose a tag to compare

What's Changed

  • feat: support anyOf in function parameters by @b0o in #13
  • Update openai types by @hmarr in #15

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

v0.2.7

12 Sep 14:20
58d7680
Compare
Choose a tag to compare

What's Changed

  • feat: add support for function_call option by @b0o in #10
  • Add debug step to log OIDC ID token by @feelepxyz in #11

New Contributors

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

v0.2.6

24 Aug 16:06
6d8b36d
Compare
Choose a tag to compare

What's Changed

  • Fix estimate for nested descriptions by @piotrkuczko in #7
  • Update OpenAI library & types

New Contributors

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

v0.2.5

14 Aug 16:00
4f87167
Compare
Choose a tag to compare

What's Changed

New Contributors

Full Changelog: v0.2.4...v0.2.5

v0.2.4

07 Aug 14:33
8c20a05
Compare
Choose a tag to compare

What's Changed

New Contributors

Full Changelog: v0.2.3...v0.2.4

v0.2.3

23 Jul 16:16
426712b
Compare
Choose a tag to compare
  • Pad first system message with newline, fixing off-by-one discrepancy

v0.2.2

23 Jul 15:29
3f6f75a
Compare
Choose a tag to compare
  • Adjust for the combination of a system message and functions (reduces token usage by 3)
  • Adjust for functions without descriptions (removes the description line altogether)

v0.2.1

23 Jul 03:07
6c5f749
Compare
Choose a tag to compare
  • Include function_call property in estimate

v0.2.0

17 Jul 20:53
5aa8bf9
Compare
Choose a tag to compare
  • Switch to pure JavaScript tokenizer (js-tiktoken)
  • Stop unnecessarily reloading the tiktoken encoding

v0.1.0

14 Jul 14:00
c8e44c2
Compare
Choose a tag to compare

Initial release