Basaran Versions Save

Basaran is an open-source alternative to the OpenAI text completion API. It provides a compatible streaming API for your Hugging Face Transformers-based text generation models.

v0.13.3

1 year ago
  • [020445c] build(*): bump version to 0.13.3 (#96)
  • [d57f4dd] docs(readme): add instructions for installing with pip (#95)
  • [bbd711a] build(deps): include playground in wheel distributions (#94)

v0.13.2

1 year ago
  • [89885a5] build(*): bump version to 0.13.2 (#93)
  • [2caee32] build(ci): upload artifacts to releases (#92)
  • [398ff8c] build(make): add targets for workflows (#91)
  • [2d50ff4] build(make): exclude virtual environment from linting (#90)
  • [0d0b2d8] build(deps): update transformers[sentencepiece] requirement (#89)
  • [f2ca237] docs(readme): update roadmap and compatibility information (#88)
  • [147a836] docs(cite): add citation file (#87)

v0.13.1

1 year ago
  • [127959e] build(*): bump version to 0.13.1 (#86)
  • [4eb9e5e] build(docker): revert base image version from 23.02-py3 to 22.12-py3 (#85)

v0.13.0

1 year ago
  • [46b86b5] build(*): bump version to 0.13.0 (#83)
  • [e5042e9] build(docker): update base image version from 22.08-py3 to 23.02-py3 (#82)
  • [3939eb1] build(deps): remove tenacity from dependencies (#81)
  • [821b9f0] fix(model): reduce memory usage while loading in half-precision (#79)
  • [506514f] build(docker): use half precision in examples (#78)
  • [8aa8443] feat(model): remove auto retrying (#77)
  • [8cab926] build(deps): update minimum requirements (#76)
  • [918373d] build(deps): update bitsandbytes requirement from ~=0.37.1 to ~=0.37.2 (#75)
  • [5741b53] docs(readme): add pypi badge (#74)
  • [f90603a] test(example): use real model in example to reduce ambiguity (#72)
  • [3fe56e8] docs(readme): update example of using as a library (#71)

v0.12.1

1 year ago
  • [42b8493] build(*): bump version to 0.12.1 (#69)
  • [ac24ba9] test(example): add example of using as a library (#68)
  • [67269c2] build(ci): remove workflow for publishing test packages (#67)
  • [04c016c] build(ci): rename codeql-analysis.yml to codeql.yml (#66)
  • [a9b84fe] build(ci): fix package publishing options (#65)
  • [c0d33e1] build(ci): add workflow for publishing packages (#64)
  • [b6dac4d] build(deps): update transformers[sentencepiece] requirement (#63)
  • [b86f061] build(deps): update huggingface-hub requirement (#60)

v0.12.0

1 year ago
  • [88d882c] build(make): bump version to 0.12.0 (#52)
  • [b417b52] build(docker): bump version to 0.12.0 (#51)
  • [6d03b98] build(docker): bundle a patched version of chatglm-6b (#55)
  • [0f3c44c] feat(model): allow to load a specific revision (#54)
  • [50521fd] build(utils): allow to download a specific revision (#53)
  • [a0bc6c3] fix(model): override torch dtype in 8-bit mode (#50)
  • [056d25a] build(docker): add example for bundling chatglm-6b (#49)
  • [9933b77] build(deps): update transformers[sentencepiece] requirement (#46)
  • [d10eb70] feat(model): add support for half-precision (#45)

v0.11.0

1 year ago

Special thanks to @volpeon for contributions in improving playground usability.


  • [8477673] build(make): bump version to 0.11.0 (#43)
  • [1db2c2f] build(docker): bump version to 0.11.0 (#42)
  • [49ad7ed] fix(playground): revert button active background color (#44)
  • [2d24eed] refactor(playground): keep only supported fields in presets (#41)
  • [9a08fe9] feat(playground): styling the stop button (#40)
  • [f56ec7f] build(deps): update accelerate requirement from ~=0.17.0 to ~=0.17.1 (#38)
  • [f7d829e] build(deps): update huggingface-hub requirement (#39)
  • [a4eb4bf] feat(playground): usability improvements (#34)
  • [b686324] feat(model): allow to trust remote code (#37)
  • [0e5630f] build(deps): update bitsandbytes requirement from ~=0.37.0 to ~=0.37.1 (#36)
  • [502fb5f] build(deps): revert huggingface-hub requirement (#33)
  • [3b63f8d] docs(readme): add instructions for running without docker (#32)
  • [a8be204] build(ci): enable dependency caching (#31)
  • [e6cc278] docs(readme): reduce verbosity (#30)
  • [ec56cf1] build(deps): update huggingface-hub requirement (#29)
  • [1a363ad] build(deps): update accelerate requirement from ~=0.16.0 to ~=0.17.0 (#28)

v0.10.0

1 year ago
  • [7b445ad] build(make): bump version to 0.10.0 (#25)
  • [34f2621] build(docker): bump version to 0.10.0 (#24)
  • [902f3bd] test(tokenizer): byte-level is not plain byte-pair-encoding (#23)
  • [b3cdb09] feat(server): add option to disable playground (#22)
  • [d1e03bc] docs(readme): add links to sample docker compose files (#21)
  • [ac0ef87] build(docker): add sample docker compose files (#20)
  • [5d12fd9] build(docker): rename bundle directory (#19)
  • [2c7eda5] docs(readme): add link to latest tags (#18)
  • [3c2a4d5] docs(readme): add tl;dr section (#17)

v0.9.0

1 year ago
  • [ba60eb3] build(make): bump version to 0.9.0
  • [32e5eb2] build(docker): bump version to 0.9.0
  • [45a93c2] docs(readme): review and tweak readme (#14)
  • [3dc5b88] refactor(playground): clean up front-end code
  • [ddd7fb5] refactor(model): minor code style updates
  • [1734206] refactor(model): tweak comments to avoid ambiguity
  • [ee3de01] refactor(tokenizer): rename decoder to tokenizer to avoid ambiguity
  • [5b72e33] refactor(model): minor code style updates
  • [aa05f3d] fix(server): rename playground handler
  • [3011dae] refactor(model): add leading underscore for internal methods
  • [0291fa5] refactor(decoder): tweak comments
  • [581a34b] test(choice): include logprobs in tests
  • [8d674c1] test(decoder): add unit tests for decoder
  • [341e1fe] test(model): add unit tests for model
  • [9d6f186] docs(readme): add codecov badge
  • [906ad1c] build(ci): add workflow for code coverage
  • [170317a] build(make): add target for code coverage
  • [831d4e1] build(deps): add coverage to dependencies
  • [cc273e1] test(data): add tiny random models for testing

v0.8.0

1 year ago
  • [14bc315] build(make): bump version to 0.8.0
  • [f8306e2] build(docker): bump version to 0.8.0
  • [1fb6d5a] docs(readme): add references and examples (#13)
  • [20cf231] build(make): remove target for building bundled images
  • [9333ea5] build(docker): add example for bundling local models
  • [09db5c1] docs(readme): better gif quality
  • [dd53837] test(example): add example of using client library
  • [77a78b2] docs(readme): add status badges
  • [cace1eb] docs(readme): limit image width
  • [5ed6517] docs(readme): convert demo video to gif
  • [8f3b005] docs(readme): add embedded video
  • [a4b240b] fix(playground): remove rich text formatting in prompt
  • [9cec10e] docs(readme): add roadmap and contributing sections
  • [9bd2b4b] build(deps): bump actions/setup-python from 3 to 4 (#10)
  • [18c8a59] build(deps): update tenacity requirement from ~=8.2.1 to ~=8.2.2 (#12)
  • [218db8e] build(ci): add workflow for testing
  • [9ba59ba] build(make): remove pycache on clean
  • [7d4f41b] test(choice): add unit tests for choice
  • [e304482] build(make): add target for testing with pytest
  • [4669249] build(make): add target for linting with flake8
  • [1c86872] build(deps): add flake8 to dependencies
  • [84c54dc] feat(playground): update default options