⚡ Performance Improvements
- **Outlines Core:** Enjoy faster FSM index construction with a new implementation (1175).
- **98% Reduction in Runtime Overhead:** Reduced overhead by storing FSM-token-mask as tensors. (1013)
🚀 New Features
- **[Transformers Vision Models](https://dottxt-ai.github.io/outlines/reference/models/transformers_vision/):** Apply structured generation with vision + text inputs (#1052)
- **[OpenAI-Compatible API Support](https://dottxt-ai.github.io/outlines/reference/models/openai/#structured-generation-support):** Use `models.openai` with any OpenAI-like API (e.g. vLLM, ChatGPT), including structured generation with `generate.json` and `generate.choice` (1142).
💡 Enhancements
- **Unified Logits Processors:** All models now use shared `outlines.processors`, completed by adding the following to the integration: [llama-cpp](https://github.com/dottxt-ai/outlines/pull/997), [vLLM](https://github.com/dottxt-ai/outlines/pull/1053) and [ExLlamaV2](https://github.com/dottxt-ai/outlines/pull/1010)).
- **Custom Regex Parsers:** Simplify the implementation of custom Guide classes with Regex Parser support (1039).
- **Qwen-style Byte Tokenizer Support:** Now compatible with Qwen-style byte tokenizers (1153).
🐛 Bug Fixes
- **CFG Beta**: Fixed large number of bugs to enable beta version grammar-based generation using Lark (1067)
- Fixed incorrect argument order breaking some models in `models.transformers_vision` (1077).
- Resolved OpenAI fallback tokenizer issue (1046).
- Option to disable tqdm bars during inference with vLLM (1004).
- `models.llamacpp` no longer includes implicit `max_tokens` (996).
- Fixed whitespace handling for `models.mlxlm` (1003).
- `models.mamba` now working, and supporting structured generation (1040).
- Resolved `pad_token_id` reset issue in `TransformerTokenizer` (1068).
- Fixed `outlines.generate` generator reuse causing runtime errors (1160).
⚠️ Breaking Changes
- `outlines.integrations` is now deprecated: 1061
Full Changeset
* Add contributors, creation and update date by rlouf in https://github.com/dottxt-ai/outlines/pull/1000
* Add SimToM prompt recipe by kstathou in https://github.com/dottxt-ai/outlines/pull/1002
* Ensure `models.llamacpp` Doesn't Have Implicit `max_tokens` by lapp0 in https://github.com/dottxt-ai/outlines/pull/996
* fix `models.mlxlm` whitespace prefix handling by lapp0 in https://github.com/dottxt-ai/outlines/pull/1003
* Adding the option to avoid displaying tqdm bars at inference with `vllm` by BIMM99 in https://github.com/dottxt-ai/outlines/pull/1004
* Update `pyproject.toml`, enable `mlx-lm` requirement only on darwin, disable `vllm` requirement on darwin by lapp0 in https://github.com/dottxt-ai/outlines/pull/1005
* Add asknews example by robcaulk in https://github.com/dottxt-ai/outlines/pull/1008
* Improve `outlines.processors`, add integration tests to test_generate.py by lapp0 in https://github.com/dottxt-ai/outlines/pull/998
* Abridged version of the .txt article on Coding For Structured Generat… by willkurt in https://github.com/dottxt-ai/outlines/pull/1012
* `RegexFSM`: Cache Legal-Token Mask as `torch.Tensor` to Improve Performance by lapp0 in https://github.com/dottxt-ai/outlines/pull/1013
* JSON response format fix by rshah713 in https://github.com/dottxt-ai/outlines/pull/1029
* Fix broken link in README.md regarding Serving with vLLM by shinohara-rin in https://github.com/dottxt-ai/outlines/pull/1030
* Major bug & fix: Fix bug in batched multi sample generation by JulesGM in https://github.com/dottxt-ai/outlines/pull/1025
* Update file contributors style for docs by gtsiolis in https://github.com/dottxt-ai/outlines/pull/1034
* fix: ocd missing blank space by talboren in https://github.com/dottxt-ai/outlines/pull/1036
* Adds support for custom regex parsers (for multimodal structured generation) by leloykun in https://github.com/dottxt-ai/outlines/pull/1039
* Update `models.transformers` to use `SequenceGeneratorAdapter` and `OutlinesLogitsProcessors` by lapp0 in https://github.com/dottxt-ai/outlines/pull/966
* Use `outlines.processors` for `models.llamacpp` by lapp0 in https://github.com/dottxt-ai/outlines/pull/997
* Add QA with Citations example to the Cookbook by alonsosilvaallende in https://github.com/dottxt-ai/outlines/pull/1042
* Fix mamba integration by making it a variant of `outlines.models.transformers` by lapp0 in https://github.com/dottxt-ai/outlines/pull/1040
* fix PyPI name for autoawq by davanstrien in https://github.com/dottxt-ai/outlines/pull/1045
* add fallback tokenizer by JerryKwan in https://github.com/dottxt-ai/outlines/pull/1046
* Update cerebrium instructions by milo157 in https://github.com/dottxt-ai/outlines/pull/1047
* Add Knowledge Graph Extraction example to the Cookbook by alonsosilvaallende in https://github.com/dottxt-ai/outlines/pull/1049
* Correct link and add llama-cpp-python installation instructions by alonsosilvaallende in https://github.com/dottxt-ai/outlines/pull/1051
* Introduce `outlines.models.transformers_vision` by lapp0 in https://github.com/dottxt-ai/outlines/pull/1052
* Use `outlines.processors` and `SequenceGeneratorAdapter` for `outlines.models.vllm` by lapp0 in https://github.com/dottxt-ai/outlines/pull/1053
* Modal documentation: fix deprecated memory= parameter by Perdu in https://github.com/dottxt-ai/outlines/pull/1057
* Documentation: Fix failing Modal example by Perdu in https://github.com/dottxt-ai/outlines/pull/1058
* Add links to the two examples by alonsosilvaallende in https://github.com/dottxt-ai/outlines/pull/1062
* Add link to Multimodal Structured Generation (MMSG) library to docs by leloykun in https://github.com/dottxt-ai/outlines/pull/1064
* Correct the documentation to disable caching by alonsosilvaallende in https://github.com/dottxt-ai/outlines/pull/1069
* Fix TransformerTokenizer pad_token_id reset by ispobock in https://github.com/dottxt-ai/outlines/pull/1068
* Fix link to `mamba` model reference by rlouf in https://github.com/dottxt-ai/outlines/pull/1072
* More detailed mlxlm documentation by lapp0 in https://github.com/dottxt-ai/outlines/pull/1074
* Add chain of thought example by alonsosilvaallende in https://github.com/dottxt-ai/outlines/pull/1087
* Fix coverage `exclude_lines` setting for ellipsis by brandonwillard in https://github.com/dottxt-ai/outlines/pull/1089
* Add ReAct agent example by alonsosilvaallende in https://github.com/dottxt-ai/outlines/pull/1090
* Include SPDX license info in project metadata by tiran in https://github.com/dottxt-ai/outlines/pull/1094
* Update Documentation by lapp0 in https://github.com/dottxt-ai/outlines/pull/1063
* Make `model_class` required arg, default `processor_class` to `AutoProcessor` by parkervg in https://github.com/dottxt-ai/outlines/pull/1077
* Fix details in the documentation by rlouf in https://github.com/dottxt-ai/outlines/pull/1096
* Change cookbook examples: Download model weights in the hub cache folder by alonsosilvaallende in https://github.com/dottxt-ai/outlines/pull/1097
* Correct variable name in chain-of-thought example by cpfiffer in https://github.com/dottxt-ai/outlines/pull/1101
* Remove deprecated `outlines.integrations` by rlouf in https://github.com/dottxt-ai/outlines/pull/1061
* Use relative coverage source paths by brandonwillard in https://github.com/dottxt-ai/outlines/pull/1113
* Add missing CI matrix step by brandonwillard in https://github.com/dottxt-ai/outlines/pull/1124
* Update modal example by cpfiffer in https://github.com/dottxt-ai/outlines/pull/1111
* docs: fix typo by billmetangmo in https://github.com/dottxt-ai/outlines/pull/1120
* Pass `text` and `images` as kwargs to VLM processor by lapp0 in https://github.com/dottxt-ai/outlines/pull/1126
* Update `CFGGuide` to use `outlines.fsm.parsing`. Enable `generate.cfg` by lapp0 in https://github.com/dottxt-ai/outlines/pull/1067
* Include hidden files in coverage CI upload by brandonwillard in https://github.com/dottxt-ai/outlines/pull/1136
* Add documentation request issue template by cpfiffer in https://github.com/dottxt-ai/outlines/pull/1138
* Set the tokenizer versions in `test_create_fsm_index_tokenizer` by brandonwillard in https://github.com/dottxt-ai/outlines/pull/1139
* Change Outlines' logo by rlouf in https://github.com/dottxt-ai/outlines/pull/1143
* Update logo size in documentation by rlouf in https://github.com/dottxt-ai/outlines/pull/1144
* Improve sampler docs by cpfiffer in https://github.com/dottxt-ai/outlines/pull/1141
* Update vllm.md by americanthinker in https://github.com/dottxt-ai/outlines/pull/1137
* Correct pathways, update site color, front page fixes by cpfiffer in https://github.com/dottxt-ai/outlines/pull/1146
* Change the color of the logo by rlouf in https://github.com/dottxt-ai/outlines/pull/1155
* Remove Broken pyairports Package, Replace with airportsdata by lapp0 in https://github.com/dottxt-ai/outlines/pull/1156
* Enable Tokenizers with Byte Tokens by lapp0 in https://github.com/dottxt-ai/outlines/pull/1153
* Integrate OpenAI API Structured Generation by lapp0 in https://github.com/dottxt-ai/outlines/pull/1142
* Remove link to Outlines twitter account by rlouf in https://github.com/dottxt-ai/outlines/pull/1168
* Don't re-use logits processors in SequenceGeneratorAdapter, copy them by lapp0 in https://github.com/dottxt-ai/outlines/pull/1160
* Fix benchmark workflow triggers by brandonwillard in https://github.com/dottxt-ai/outlines/pull/1170
* Reuse jinja environment for a prompt by jantrienes in https://github.com/dottxt-ai/outlines/pull/1162
* Use Faster FSM by lapp0 in https://github.com/dottxt-ai/outlines/pull/1175
* Pin `outlines-core` version by brandonwillard in https://github.com/dottxt-ai/outlines/pull/1187
* add missing comma in llamacpp docs by cpfiffer in https://github.com/dottxt-ai/outlines/pull/1190
* Minor Changes on Top of Isamus EXL2 PR (Docs, Enable FSM/CFG, Fix Tokenizer) by lapp0 in https://github.com/dottxt-ai/outlines/pull/1191
* Version the Documentation by lapp0 in https://github.com/dottxt-ai/outlines/pull/1059
New Contributors
* BIMM99 made their first contribution in https://github.com/dottxt-ai/outlines/pull/1004
* robcaulk made their first contribution in https://github.com/dottxt-ai/outlines/pull/1008
* willkurt made their first contribution in https://github.com/dottxt-ai/outlines/pull/1012
* rshah713 made their first contribution in https://github.com/dottxt-ai/outlines/pull/1029
* shinohara-rin made their first contribution in https://github.com/dottxt-ai/outlines/pull/1030
* JulesGM made their first contribution in https://github.com/dottxt-ai/outlines/pull/1025
* gtsiolis made their first contribution in https://github.com/dottxt-ai/outlines/pull/1034
* talboren made their first contribution in https://github.com/dottxt-ai/outlines/pull/1036
* JerryKwan made their first contribution in https://github.com/dottxt-ai/outlines/pull/1046
* milo157 made their first contribution in https://github.com/dottxt-ai/outlines/pull/1047
* Perdu made their first contribution in https://github.com/dottxt-ai/outlines/pull/1057
* ispobock made their first contribution in https://github.com/dottxt-ai/outlines/pull/1068
* tiran made their first contribution in https://github.com/dottxt-ai/outlines/pull/1094
* cpfiffer made their first contribution in https://github.com/dottxt-ai/outlines/pull/1101
* billmetangmo made their first contribution in https://github.com/dottxt-ai/outlines/pull/1120
* americanthinker made their first contribution in https://github.com/dottxt-ai/outlines/pull/1137
* jantrienes made their first contribution in https://github.com/dottxt-ai/outlines/pull/1162
**Full Changelog**: https://github.com/dottxt-ai/outlines/compare/0.0.46...0.1.0