Changelog¶
All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
[0.95.0]¶
- Module changes:
- Anthropic
Invoke-AnthropicModel- Added support for two new models:
anthropic.claude-opus-4-20250514-v1:0,anthropic.claude-sonnet-4-20250514-v1:0
- Added support for two new models:
- Anthropic
[0.90.0]¶
- Module changes:
- Meta
Invoke-MetaModel- Added support for two new models:
meta.llama4-maverick-17b-instruct-v1:0,meta.llama4-scout-17b-instruct-v1:0
- Added support for two new models:
- Meta
[0.85.0] - BREAKING CHANGES¶
- Module changes:
- Converse
Invoke-ConverseAPI- parameter
MediaPathrenamed toImagePath- Breaking Change - parameter
AdditionalModelRequestFieldchange from typePSObjectto[System.Collections.Specialized.OrderedDictionary]- Breaking Change - Fixed bug where use of
AdditionalModelRequestFieldwas not being properly formatted - Added new parameters
VideoPath,S3Location, andS3BucketOwnerto support Vision models that can take video files
- parameter
- Updated model pricing to latest published model prices
- Minor spelling corrections throughout
- Converse
[0.75.0]¶
- Module changes:
- Mistral AI
Invoke-MistralAIChatModel.ps1- Added support for new model:
mistral.pixtral-large-2502-v1:0- Added new
MediaPathparameter as this new model supports vision
- Added new
- Fixed bug where
mistral.mistral-small-2402-v1:0was not functioning properly.
- Added support for new model:
- Mistral AI
[0.70.0]¶
- Module changes:
- Amazon
- Added new function
Invoke-AmazonVideoModelto support text model:amazon.nova-reel-v1:1
- Added new function
- Amazon
[0.65.0]¶
- Module changes:
- DeepSeek
- Added new function
Invoke-DeepSeekModelto support text model:deepseek.r1-v1:0
- Added new function
- DeepSeek
[0.60.0]¶
- Module changes:
- Anthropic
Invoke-AnthropicModel- added support for new model:
anthropic.claude-3-7-sonnet-20250219-v1:0 - added new
ThinkingandThinkingBudgetTokensparameters to support thinking and reasoning outputs for Claude 3.7 Sonnet
- added support for new model:
- Anthropic
[0.55.0]¶
- Module changes:
- Luma
- Added new function
Invoke-LumaAIModelto support video model:luma.ray-v2:0
- Added new function
- Manifest updates:
- Require more recent version of
AWS.Tools.BedrockRuntimeto take advantage of new commands such asStart-BDRRAsyncInvoke - Require
AWS.Tools.S3to support async video file generation download from S3.
- Require more recent version of
- Minor syntax formatting corrections throughout to address new PSScriptAnalyzer rules
- Luma
- Build changes:
- Updated bootstrap file to latest versions of modules
- Updated all GitHub actions workflows from
actions/upload-artifact@v3toactions/upload-artifact@v4 pwshBedrock.build.ps1Improvements:Invoke-Formatternow runs at the end ofBuildstep for final cleanup- Added new Markdown processing functionality to address two Markdown linting issues:
MD022/blanks-around-headingsMD040/fenced-code-language
[0.52.0] - BREAKING CHANGES¶
- Module changes:
- Added two new properties to all model documentation objects:
Image- designates if the model can generate imagesVideo- designates if the model can generate videoVision- re-purposed to now designate if the model can see and describe uploaded content- This required some logic re-work in several functions that were previously using
Visionto designate image capability
- This required some logic re-work in several functions that were previously using
- Added inference profile ID support to all models IAW Supported cross-region inference profiles
- pwshBedrock will now properly format the inference profile ID for all models that support which enables the ability to make Bedrock calls to other regions
Reset-ModelContext- fixed issue where certain models were missing from validation.- Amazon
Invoke-AmazonNovaTextModel- added support for new Nova Text models:amazon.nova-pro-v1:0,amazon.nova-lite-v1:0,amazon.nova-micro-v1:0Invoke-AmazonImageModel- added support for new Nova image model:
amazon.nova-canvas-v1:0 - fixed bug where files would not always save to drive
CfgScaleparameter maximum adjusted from 10 to 9.9
- added support for new Nova image model:
- Meta
- Removed
meta.llama2-13b-chat-v1andmeta.llama2-70b-chat-v1from all functions as Bedrock has EOL these two models. - Breaking Change - Updated
Invoke-MetaModelto support new model:meta.llama3-3-70b-instruct-v1:0
- Removed
- stability.ai
- Updated
Invoke-StabilityAIImageModelto support new model:stability.sd3-5-large-v1:0
- Updated
- Updated pricing for all models
- Added two new properties to all model documentation objects:
- Build changes:
- Updated bootstrap file to latest versions of modules
[0.33.0] - BREAKING CHANGES¶
- Module changes:
- stability.ai
Invoke-StabilityAIDiffusionModelrenamed toInvoke-StabilityAIDiffusionXLModel- Breaking Change- Updated help which had incorrect examples
Heightparameter now defaults to 1024 if no user input is provided.Widthparameter now defaults to 1024 if no user input is provided.
- Added
Invoke-StabilityAIImageModelto support new models:stability.stable-image-core-v1:0,stability.stable-image-ultra-v1:0,stability.sd3-large-v1:0
- AI21 Labs
Invoke-AI21LabsJambaModel- added support forai21.jamba-1-5-mini-v1:0andai21.jamba-1-5-large-v1:0Invoke-AI21LabsJurassic2Modelremoved. All support forai21.j2-grande-instruct,ai21.j2-jumbo-instruct,ai21.j2-mid-v1,ai21.j2-ultra-v1removed. These models are being EOL in Amazon Bedrock. - Breaking Change
- Amazon
Invoke-AmazonTextModel- minor corrections to debug output and help
- Cohere
Invoke-CohereCommandRModel- minor corrections to debug output
- Meta
Invoke-MetaModel- Added support for new 3.2 models:
meta.llama3-2-1b-instruct-v1:0,meta.llama3-2-3b-instruct-v1:0,meta.llama3-2-11b-instruct-v1:0,meta.llama3-2-90b-instruct-v1:0 - Added vision support for new large 3.2 models
- Added tools support for all Meta models 3.1 and higher
- minor corrections to debug output and help
- Added support for new 3.2 models:
- Mistral
Invoke-MistralAIModel- minor corrections to debug output and help. Adjusted Max token limit validation.
- Anthropic
Invoke-AnthropicModel- added support for new models:anthropic.claude-3-5-haiku-20241022-v1:0,anthropic.claude-3-5-sonnet-20241022-v2:0
- Updated pricing for all models
- stability.ai
- Build changes:
- Updated bootstrap AWS modules from
4.1.621to4.1.700 - Switched
wf_Linuxandwf_Windowsfrom CodeBuild actions to GitHub Actions runners due to performance issues with CB.
- Updated bootstrap AWS modules from
[0.15.0]¶
- Module changes:
- Amazon
- Added support for Titan Image Generator G1 V2 -
amazon.titan-image-generator-v2:0- Added new Conditioned Image Generation parameters to
Invoke-AmazonImageModel - Added new Color Guided Content parameters to
Invoke-AmazonImageModel - Added new Background removal parameters to
Invoke-AmazonImageModel
- Added new Conditioned Image Generation parameters to
- Added support for Titan Image Generator G1 V2 -
- Meta
- Added support for Llama 3.1 405B Instruct -
meta.llama3-1-405b-instruct-v1:0 - Updated pricing to reflect current Meta Llama 3.1 prices
- Corrected Meta Llama 3.1 models to show Multilingual support as
$true
- Added support for Llama 3.1 405B Instruct -
- Mistral AI
- Updated pricing to reflect current prices
- Amazon
- Build changes:
- Updated tests to follow Pester 5 rules
- Updated integration tests for newly added capabilities
[0.9.1]¶
- Updated IconUri in manifest.
[0.9.0]¶
Added¶
- Initial release.