Release v0.2.7

What's Changed

• improve tests by @aniketmaurya in #418
• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci in #421
• git: Handle None values in ChatMessage content by @Lucaz0619 in #422
• fix custom exceptions by @aniketmaurya in #425
• fix: reponse format JSONSchema key fix by @Lucaz0619 in #427
• fix: ensure proper cleanup of processes in wrap_litserve_start by @bhimrazy in #432
• fix async continuous batching by @aniketmaurya in #429
• [fix] continuous batching - fix prefill by @ali-alshaar7 in #433
• Feat: add custom health check logic by @bhimrazy in #430
• fix: Replace deprecated dict method with model_dump in OpenAI Spec decode step by @bhimrazy in #434
• fix: update type hints for _default_unbatch and _spec attributes in LitAPI class by @bhimrazy in #435
• Feature: 🚀 Add Audio Content Support to OpenAISpec Request by @bhimrazy in #439
• hotfix: pin starlette dependency by @bhimrazy in #445
• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci in #442
• feat: add lightning cli dynamically by @aniketmaurya in #446
• Release 0.2.7 by @aniketmaurya in #447

New Contributors

• @Lucaz0619 made their first contribution in #422

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

v0.2.7.dev0

What's Changed

• improve test timeouts by @aniketmaurya in #418
• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci in #421
• git: Handle None values in ChatMessage content by @Lucaz0619 in #422
• fix custom exceptions by @aniketmaurya in #425
• fix: reponse format JSONSchema key fix by @Lucaz0619 in #427
• fix: ensure proper cleanup of processes in wrap_litserve_start by @bhimrazy in #432
• fix async continuous batching by @aniketmaurya in #429
• [fix] continuous batching - fix prefill by @ali-alshaar7 in #433
• Feat: add custom health check logic by @bhimrazy in #430

New Contributors

• @Lucaz0619 made their first contribution in #422

Full Changelog: v0.2.6...v0.2.7.dev0

v0.2.6

What's Changed

• feat: info route by @lorenzomassimiani in #368
• Fix CI: async tests with ASGITransport by @aniketmaurya in #376
• Fix: Replace Deprecated max_tokens with max_completion_tokens in OpenAI Spec by @rittik9 in #375
• feat: Customizable Loops 1/n by @aniketmaurya in #374
• customizable loop - wire up Loops to LitServer 2/n by @aniketmaurya in #378
• Improve CI: retry flaky tests by @aniketmaurya in #379
• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci in #377
• check device format while initialising litserver by @ali-alshaar7 in #380
• Release 0.2.6.dev0 by @aniketmaurya in #383
• Update PR template by @rittik9 in #381
• Include user field and base64 literal for encoding_format by @aniketmaurya in #388
• Improve error handling and debugging experience by @aniketmaurya in #389
• improved logging with sensible defaults by @aniketmaurya in #391
• add continuous batching loop 1/n by @aniketmaurya in #387
• Add loop.pre_setup to allow fine-grained LitAPI validation based on inference loop by @aniketmaurya in #393
• Make LitAPI.predict optional and validate API implementation by @aniketmaurya in #394
• Fix OpenAISpec with continuous batching loop by @aniketmaurya in #395
• add tests for continuous batching and Default loops by @aniketmaurya in #396
• Set LitServer.stream using LitSpec.stream by @aniketmaurya in #398
• fix openai usage info for non-streaming response by @aniketmaurya in #399
• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci in #400
• Async continuous batching loop by @aniketmaurya in #401
• add validation for stream=False with yield usage by @aniketmaurya in #402
• fix callback runner to execute after predict by @aniketmaurya in #406
• integrate zmq by @aniketmaurya in #403
• warn users when predict/unbatch output length is not same as #requests by @aniketmaurya in #408
• move built in loops inside classes by @aniketmaurya in #409
• add justus and thomas as codeowners by @aniketmaurya in #410
• enable multiple workers for ZMQ by @aniketmaurya in #411
• Fix: Add Callback Events and Align Hooks in Streaming Loop by @bhimrazy in #407
• bump: Lightning-AI/utilities used main by @Borda in #415
• Release v0.2.6 by @aniketmaurya in #413
• fix: don't start zmq when fast_queue=false by @aniketmaurya in #417
• fix release ci by @aniketmaurya in #416

New Contributors

• @rittik9 made their first contribution in #375
• @ali-alshaar7 made their first contribution in #380

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

v0.2.6.dev3

What's Changed

• Set LitServer.stream using LitSpec.stream by @aniketmaurya in #398
• fix openai usage info for non-streaming response by @aniketmaurya in #399
• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci in #400
• Async continuous batching loop by @aniketmaurya in #401
• add validation for stream=False with yield usage by @aniketmaurya in #402
• fix callback runner to execute after predict by @aniketmaurya in #406
• integrate zmq by @aniketmaurya in #403
• warn users when predict/unbatch output length is not same as #requests by @aniketmaurya in #408
• move built in loops inside classes by @aniketmaurya in #409
• add justus and thomas as codeowners by @aniketmaurya in #410
• enable multiple workers for ZMQ by @aniketmaurya in #411

Full Changelog: v0.2.6.dev2...v0.2.6.dev3

v0.2.6.dev2

What's Changed

• Make LitAPI.predict optional and validate API implementation by @aniketmaurya in #394
• Fix OpenAISpec with continuous batching loop by @aniketmaurya in #395
• add tests for continuous batching and Default loops by @aniketmaurya in #396

Full Changelog: v0.2.6.dev1...v0.2.6.dev2

v0.2.6.dev1

What's Changed

• Update PR template by @rittik9 in #381
• Include user field and base64 literal for encoding_format by @aniketmaurya in #388
• Improve error handling and debugging experience by @aniketmaurya in #389
• improved logging with sensible defaults by @aniketmaurya in #391
• add continuous batching loop 1/n by @aniketmaurya in #387
• Add loop.pre_setup to allow fine-grained LitAPI validation based on inference loop by @aniketmaurya in #393

Full Changelog: v0.2.6.dev0...v0.2.6.dev1

v0.2.6.dev0

What's Changed

• feat: info route by @lorenzomassimiani in #368
• Fix CI: async tests with ASGITransport by @aniketmaurya in #376
• Fix: Replace Deprecated max_tokens with max_completion_tokens in OpenAI Spec by @rittik9 in #375
• feat: Customizable Loops 1/n by @aniketmaurya in #374
• customizable loop - wire up Loops to LitServer 2/n by @aniketmaurya in #378
• Improve CI: retry flaky tests by @aniketmaurya in #379
• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci in #377
• check device format while initialising litserver by @ali-alshaar7 in #380
• Release 0.2.6.dev0 by @aniketmaurya in #383

New Contributors

• @rittik9 made their first contribution in #375
• @ali-alshaar7 made their first contribution in #380

Full Changelog: v0.2.5...v0.2.6.dev0

v0.2.5

What's Changed

• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci in #353
• Fail fast when LitAPI.setup has error by @aniketmaurya in #356
• Development release 0.2.5.dev0 by @aniketmaurya in #357
• make @Borda global code owner by @aniketmaurya in #358
• Update media type and JSON handling in OpenAISpec by @fabigr8 in #360
• Add custom healthcheck path by @akuma12 in #342
• Automatic Serialization/deserialization of images by @aniketmaurya in #361
• Update docker base image to avoid issue with builds by @AlexShukhman in #363
• Enable multiple image input by @aniketmaurya in #364
• Test multiple Python version compatibility by @aniketmaurya in #336
• ci: prune configuration by @Borda in #365
• Feat support OpenAI embedding by @bhimrazy in #367
• Fix: Make OpenAIEmbedding work when token usage info is not set by @aniketmaurya in #369
• Assert batch size to strictly test OpenAIEmbeddingSpec for dynamic batching by @aniketmaurya in #370
• Release v0.2.5 by @aniketmaurya in #371

New Contributors

• @fabigr8 made their first contribution in #360
• @akuma12 made their first contribution in #342
• @AlexShukhman made their first contribution in #363

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

v0.2.5.dev0

What's Changed

• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci in #353
• Fail fast when LitAPI.setup has error by @aniketmaurya in #356
• Development release 0.2.5.dev0 by @aniketmaurya in #357
• make @Borda global code owner by @aniketmaurya in #358

Full Changelog: v0.2.4...v0.2.5.dev0

v0.2.4

What's Changed

• Dynamically grab port number by @grantgasser in #307
• test generate client file by @aniketmaurya in #322
• avoid client generation if already exist by @aniketmaurya in #324
• move middlewares to a module by @aniketmaurya in #326
• enable isort by @aniketmaurya in #327
• Feat: track number of active requests by @aniketmaurya in #325
• aggregate requests across uvicorn processes by @aniketmaurya in #328
• lint: emergency docformatter bump by @aniketmaurya in #331
• update(lint): prune config by @Borda in #332
• Avoiding counting active requests in /health and /metrics by @vrdn-23 in #334
• Host param by @grumpyp in #335
• improve error handling from inference worker processes by @aniketmaurya in #337
• Fix Docker Argument Handling by @aniketmaurya in #338
• Add dockerization to feature list by @aniketmaurya in #346
• stream tool calling by @whisper-bye in #349
• Release 0.2.4 by @aniketmaurya in #351

New Contributors

• @grantgasser made their first contribution in #307
• @vrdn-23 made their first contribution in #334
• @whisper-bye made their first contribution in #349

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