Outlines v0.0.31

What's Changed

• Keep track of state in RegexLogitsProcessor using input_ids by @lapp0 in #628
• Put ancestors on the same device as the KV cache by @raphaelchinchilla in #660

New Contributors

• @raphaelchinchilla made their first contribution in #660

Full Changelog: 0.0.30...0.0.31

Outlines v0.0.30

What's Changed

• Ensure Ancestors on Correct Device During Sampling by @lapp0 in #651

Full Changelog: 0.0.29...0.0.30

Outlines v0.29 〰️

What's Changed

• Make generate.format return the corresponding object by @rlouf in #630
• Add top-k processor to multinomial sampler by @rlouf in #642
• Add top_p logits processor and temperature rescaling to the multinomial sampler by @rlouf in #646

Full Changelog: 0.0.28...0.0.29

Outlines v0.0.28

What's Changed

• Use Better Models, Better Prompts in Docs by @lapp0 in #619
• Add Beam Search sampler by @rlouf in #618
• Remove deprecated interface and arguments by @rlouf in #627
• Allow JSON Schema Constraining to a Single Whitespace by @lapp0 in #625

Full Changelog: 0.0.27...0.0.28

Outlines v0.0.27

What's Changed

• Update openai_compatible.py by @7flash in #600
• allow generation of %ignore terminals by @lapp0 in #570
• Allow to generate several samples for each prompt by @rlouf in #533
• Bump torch and transformers versions by @lapp0 in #613

Full Changelog: 0.0.26...0.0.27

Outlines v0.0.26

What's Changed

• Update index.md - fix model by @7flash in #601
• Fix typo in prompt example by @ShukantPal in #603
• Reset lark package in test_parsing.py by @lapp0 in #596
• introduce WHITESPACE regex constant by @jc-louis in #599
• Make json array generation more flexible by @posionus in #608
• Update self_consistency.py - print actual answer by @7flash in #602
• Bump vLLM version to 0.3.0 by @lapp0 in #579

New Contributors

• @7flash made their first contribution in #601
• @ShukantPal made their first contribution in #603
• @jc-louis made their first contribution in #599
• @posionus made their first contribution in #608

Full Changelog: 0.0.25...0.0.26

v0.0.25

What's Changed

• Fix bug in the treatment of the anyOf keywork in JSON schema by @RobinPicard in #552
• Improvements to documentation and README by @rlouf in #547
• Added subclass of OpenAI class to support custom endpoints by @shannonsands in #560
• Update pydantic install instructions for vLLM by @rlouf in #564
• Pin transformers version to 4.36.2 by @lapp0 in #569
• Clear cache upon version upgrade by @lapp0 in #566
• Implement JSON schema field constraints by @aschwa in #567
• Fix incompleteness of regex and cfg guided generation by @benlipkin in #544
• Add benchmark suite by @lapp0 in #542
• Fix Bug Issue Template - Command Doesn't Produce Version, Isn't Copyable by @lapp0 in #578
• Bugfix: Fixes dtype issue described in #583 by @dnhkng in #584
• Make OpenAI models compatible with the main API by @rlouf in #574
• Add grammars module by @lapp0 in #562

New Contributors

• @shannonsands made their first contribution in #560
• @lapp0 made their first contribution in #569
• @aschwa made their first contribution in #567
• @dnhkng made their first contribution in #584

Full Changelog: 0.0.24...0.0.25

Outlines v0.0.24

What's Changed

• Add roadmap blog post by @rlouf in #519
• Remove draft status of roadmap post by @rlouf in #525
• Cache creation of states_token_mapping in the init of RegexFSM by @RobinPicard in #492
• Remove the AutoAWQ and AutoGPTQ integrations by @rlouf in #527
• Fix typos by @pitmonticone in #530
• Add CFG to vllm serving by @mory91 in #517
• Revert "Add CFG to vllm serving" by @rlouf in #537

New Contributors

• @pitmonticone made their first contribution in #530
• @mory91 made their first contribution in #517

Full Changelog: 0.0.23...0.0.24

Outlines v0.0.23

What's Changed

• Support required keyword in JSON Schema by @RobinPicard in #495
• Use the OpenAI API to check that a model exists by @rlouf in #518
• Updates chess cookbook example with Phi-2 and adds omitted code. by @kgourgou in #521
• Patch nest_asyncio in notebooks only by @dzhulgakov in #520
• Support Azure OpenAI by @joeddav in #522

New Contributors

• @kgourgou made their first contribution in #521
• @dzhulgakov made their first contribution in #520
• @joeddav made their first contribution in #522

Full Changelog: 0.0.22...0.0.23

Outlines v0.0.22

What's Changed

• Remove idx arg from FSM interface by @benlipkin in #510
• Add LlamaCpp integration by @dtiarks in #486

New Contributors

• @dtiarks made their first contribution in #486

Full Changelog: 0.0.21...0.0.22