0.14.0

What's Changed

• feat: add llamacpp by @njbrake in #190
• fix(Groq): Remove unsupported params from integration tests by @besaleli in #188
• fix(cohere): Introduce _patch_messages function to process tool messages by @njbrake in #195
• docs: How To Contribute A New Provider by @besaleli in #197
• chore: add codecov.yml by @besaleli in #198
• feat: Update BaseProvider to async-first implementation by @besaleli in #223
• chore(deps): Bump "mistralai>=1.9.3". by @daavoo in #238
• feat(async): Migrate OpenAI provider to async by @besaleli in #230
• feat(async): Migrate mistral provider to async by @besaleli in #231
• feat(async): Migrate Groq provider to async by @besaleli in #232
• feat(async): Migrate Anthropic provider to async by @besaleli in #234
• feat(async): Migrate cohere provider to async by @besaleli in #235
• fix(api): Set default reasoning_effort="medium". by @daavoo in #239
• feat(async): Migrate deepseek provider to async by @besaleli in #240
• fix(groq): API dies when given reasoning-effort by @besaleli in #244
• chore(deps): bump actions/checkout from 4 to 5 by @dependabot[bot] in #237
• List Available Models by @inventivepotter in #228
• feat(google): Enable reasoning. by @daavoo in #243
• feat(async): Migrate Ollama provider to async by @besaleli in #245
• feat(async): Migrate voyage provider to async by @besaleli in #246
• refactor(anthropic): Extract _convert_params. by @daavoo in #251
• feat: Update model parameter handling in API and documentation by @njbrake in #249
• feat: reasoning_effort -> map to provider defaults by @njbrake in #250
• fix(ollama): put ollama utils in try catch block by @njbrake in #256
• chore(anthropic): Increase DEFAULT_MAX_TOKENS from 4096 to 8192. by @daavoo in #260
• feat(anthropic): Enable reasoning. Handle reasoning_effort. by @daavoo in #252
• fix(google): Handle reasoning_effort defaults. by @daavoo in #261

New Contributors

• @inventivepotter made their first contribution in #228

Full Changelog: 0.13.1...0.14.0

0.13.1

What's Changed

• feat: integration testing for multiturn tool usage, patch mistral handling of tool messages by @njbrake in #187

Full Changelog: 0.13.0...0.13.1

0.13.0

What's Changed

• pre-commit autoupdate by @github-actions[bot] in #178
• fix(huggingface): Handle config.api_base. by @daavoo in #179
• tests: Enable huggingface integration tests. by @daavoo in #181
• chore: Update issue templates by @besaleli in #180
• feat(api): Update message handling in completion functions to support… by @njbrake in #174
• fix(huggingface): Use openai's type_to_response_format_param to han… by @daavoo in #185

Full Changelog: 0.12.1...0.13.0

0.12.1

What's Changed

• refactor(tests): Remove unhelpful comments from integration and unit … by @njbrake in #166
• feat: Add Llamafile provider and update dependencies by @njbrake in #167
• feat(tools): Enhance callable_to_tool to support complex types by @njbrake in #173

Full Changelog: 0.12.0...0.12.1

0.12.0

What's Changed

• feat: response_format for cohere provider by @besaleli in #157
• ci: PR template by @besaleli in #156
• fix(google): Handle multiple input messages appropriately. by @daavoo in #159
• fix(google): Extract and pass system_instruction. by @daavoo in #160
• chore: pydantic typed between api.completion <-> provider.completion by @njbrake in #154
• feat: Evaluate and fix arguments to responses API by @besaleli in #161
• fix(xai): Preserve tool_call id from client response. by @daavoo in #163
• feat: Clean up official completion parameters by @njbrake in #164
• fix(xai): Handle messages inside agent loop. by @daavoo in #165

Full Changelog: 0.11.2...0.12.0

0.11.2

What's Changed

• fix(aws): Prevent stream option with response_format (#151) by @njbrake in #153

Full Changelog: 0.11.1...0.11.2

0.11.1

What's Changed

• refactor: Better typing of ProviderMetadata and intelligent test skip by @njbrake in #136
• feat: Add typing for responses API by @besaleli in #138
• chore: Add CONTRIBUTING.md. by @daavoo in #141
• fix(anthropic): Handle messages inside agent loop. by @daavoo in #145
• tests: Add tests-docs workflow. by @daavoo in #147
• fix: cleaner imports by @njbrake in #148
• feat: Add support for VoyageAI. by @daavoo in #98
• fix: Handle unsupported stream option in response format by @njbrake in #151

Full Changelog: 0.11.0...0.11.1

0.11.0

What's Changed

• Update fireworks.py by @njbrake in #129
• feat(xai): add XAI provider with completion support by @njbrake in #128
• feat: Add Databricks Provider by @besaleli in #130
• feat(fireworks): Responses API for Fireworks AI by @Shiva4113 in #121
• refactor(ollama): Simplify response handling by consolidating respons… by @njbrake in #133

New Contributors

• @besaleli made their first contribution in #130
• @Shiva4113 made their first contribution in #121

Full Changelog: 0.10.0...0.11.0

0.10.0

What's Changed

• fix(google): Handle tool_choice. by @daavoo in #113
• chore: consistent exception raising by @njbrake in #114
• chore(deps): bump pre-commit from 4.2.0 to 4.3.0 by @dependabot[bot] in #118
• pre-commit autoupdate by @github-actions[bot] in #115
• feat(groq): Enable responses support and implement responses API method by @njbrake in #117
• chore: remove verify_kwargs by @njbrake in #119
• refactor: remove helpers and streamline response conversion by @njbrake in #124
• chore: Enable more ruff rules (list taken from any-agnet) and fix err… by @daavoo in #123
• chore: Update integration tests configuration and provider model mapping by @njbrake in #125
• chore: clean up naming by @njbrake in #126

New Contributors

• @dependabot[bot] made their first contribution in #118

Full Changelog: 0.9.0...0.10.0

0.9.0

What's Changed

• fix(anthropic): Drop duplicated stream + response_format check. by @daavoo in #106
• pre-commit autoupdate by @github-actions[bot] in #105
• fix(anthropic): Handle tool_choice=required conversion. by @daavoo in #107
• feat: Add initial responses api by @njbrake in #108
• feat: Responses API documentation and implementation by @njbrake in #109

Full Changelog: 0.8.1...0.9.0