v0.3.0

Lots of changes that includes the RC releases as well.

What's Changed

• fix openai content part media by @brainlid in #112
• ContentPart image media option updates by @brainlid in #113
• Updates for ContentPart images with messages to support ChatGPT's "detail" level option by @brainlid in #114
• add openai image endpoint support (aka DALL-E-2 & DALL-E-3) by @brainlid in #116
• allow PromptTemplates to convert to ContentParts by @brainlid in #117
• Fix elixir 1.17 warnings by @MrYawe in #123
• updates to README by @petrus-jvrensburg in #125
• Add ChatVertexAI by @raulchedrese in #124
• Major update. Preparing for v0.3.0-rc.0 - breaking changes by @brainlid in #131
• update calculator tool by @brainlid in #132
• support receiving rate limit info by @brainlid in #133
• upgrade abacus dep by @brainlid in #134
• add support for TokenUsage through callbacks by @brainlid in #137
• Big update - RC ready by @brainlid in #138
• Improvements to docs by @brainlid in #145
• ChatGoogleAI fixes and updates by @brainlid in #152
• fix: typespec error on Message.new_user/1 by @bwan-nan in #151
• Convert to use mimic for mocking calls by @brainlid in #155
• Remove ApiOverride reference in mix.exs project.docs by @stevehodgkiss in #157
• Fix OpenAI chat stream hanging by @stevehodgkiss in #156
• Fix streaming error when using Azure OpenAI Service by @stevehodgkiss in #158
• Update Azure OpenAI Service streaming fix by @stevehodgkiss in #161
• Fix ChatOllamaAI streaming response by @alappe in #162
• Fix PromptTemplate example by @joelpaulkoch in #167
• adds OpenAI project authentication. by @fbettag in #166
• Anthropic support for streamed tool calls with parameters by @brainlid in #169
• change return of LLMChain.run/2 - breaking change by @brainlid in #170
• 🐛 cast tool_calls arguments correctly inside message_deltas by @rparcus in #175
• Do not duplicate tool call parameters if they are identical by @michalwarda in #174
• Structured Outputs by supplying strict: true in #173
• feat: add OpenAI's new structured output API by @monotykamary in #180
• Support system instructions for Google AI by @elliotb in #182
• Handle empty text parts from GoogleAI responses by @elliotb in #181
• Handle missing token usage fields for Google AI by @elliotb in #184
• Handle functions with no parameters for Google AI by @elliotb in #183
• Add AWS Bedrock support to ChatAnthropic by @stevehodgkiss in #154
• Handle all possible finishReasons for ChatGoogleAI by @elliotb in #188
• Remove unused assignment from ChatGoogleAI by @elliotb in #187
• Add support for passing safety settings to Google AI by @elliotb in #186
• Add tool_choice for OpenAI and Anthropic by @avergin in #142
• add support for examples to title chain by @brainlid in #191
• add "processed_content" to ToolResult struct and support storing Elixir data from function results by @brainlid in #192
• Revamped error handling and handles Anthropic's "overload_error" by @brainlid in #194
• Documenting AWS Bedrock support with Anthropic Claude by @brainlid in #195
• Cancel a message delta when we receive "overloaded" error by @brainlid in #196
• implement initial support for fallbacks by @brainlid in #207
• Fix content-part encoding and decoding for Google API. by @vkryukov in #212
• Fix specs and examples by @vkryukov in #211
• Ability to Summarize an LLM Conversation by @brainlid in #216
• Prepare for v0.3.0-rc.1 by @brainlid in #217
• add explicit message support in summarizer by @brainlid in #220
• Change abacus to optional dep by @nallwhy in #223
• Remove constraint of alternating user, assistant by @GenericJam in #222
• Breaking change: consolidate LLM callback functions by @brainlid in #228
• feat: Enable :inet6 for Req.new for Ollama by @mpope9 in #227
• fix: enable verbose_deltas in #197
• Prep for v0.3.0-rc.2 - update version and docs outline by @brainlid in #229
• Add Bumblebee Phi-4 by @marcnnn in #233
• feat: apply chat template from callback by @joelpaulkoch in #231
• support for o1 OpenAI model by @brainlid in #234
• feat: Support for Ollama keep_alive API parameter by @mpope9 in #237
• Add prompt caching support for Claude. by @montebrown in #226
• Add raw field to TokenUsage by @nallwhy in #236
• Add LLAMA 3.1 Json tool call with Bumblebee by @marcnnn in #198
• prep for v0.3.0 release by @brainlid in #238

New Contributors

• @MrYawe made their first contribution in #123
• @petrus-jvrensburg made their first contribution in #125
• @raulchedrese made their first contribution in #124
• @bwan-nan made their first contribution in #151
• @stevehodgkiss made their first contribution in #157
• @alappe made their first contribution in #162
• @joelpaulkoch made their first contribution in #167
• @fbettag made their first contribution in #166
• @rparcus made their first contribution in #175
• @monotykamary made their first contribution in #180
• @elliotb made their first contribution in #182
• @avergin made their first contribution in #142
• @vkryukov made their first contribution in #212
• @nallwhy made their first contribution in #223
• @GenericJam made their first contribution in #222
• @mpope9 made their first contribution in #227
• @marcnnn made their first contribution in #233
• @montebrown made their first contribution in #226

Full Changelog: v0.2.0...v0.3.0

Thanks to all the contributors!