Adding more bedrock Claude models with Cross-Region Inference

[brunobpr]

Adding the following models:
Claude 3.5 Sonnet v2:

• us.anthropic.claude-3-5-sonnet-20241022-v2:0
• apac.anthropic.claude-3-5-sonnet-20241022-v2:0

Claude 3.5 Haiku v1 :

• us.anthropic.claude-3-5-haiku-20241022-v1:0

Claude 3.7 Sonnet v1:

• us.anthropic.claude-3-7-sonnet-20250219-v1:0

---

Important

Add new Claude models for cross-region inference to _supported__models in claude.py.

• Models:
  • Add us.anthropic.claude-3-5-sonnet-20241022-v2:0, apac.anthropic.claude-3-5-sonnet-20241022-v2:0, us.anthropic.claude-3-5-haiku-20241022-v1:0, and us.anthropic.claude-3-7-sonnet-20250219-v1:0 to _supported__models in claude.py.

^{This description was created by for 75b3f8a. It will automatically update as commits are pushed.}

[ellipsis-dev]

👍 Looks good to me! Reviewed everything up to 75b3f8a in 1 minute and 24 seconds

More details

• Looked at 15 lines of code in 1 files
• Skipped 0 files when reviewing.
• Skipped posting 2 drafted comments based on config settings.

1. extensions/llms/bedrock/pandasai_bedrock/claude.py:37

• Draft comment:
  New model identifiers added. Ensure corresponding unit tests and documentation updates exist for these new models.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50%
  The comment asks the author to "ensure" something, which violates the rule about not asking authors to verify things. Adding model strings to a list is a very simple change that likely doesn't require additional tests - the existing tests would cover the basic functionality. The models are just strings used for validation. The docstring purposely doesn't list specific models since they change frequently.
  Perhaps these new models have special handling requirements or different response formats that do need testing? Maybe the documentation should track supported model versions?
  Looking at the code, all models are handled exactly the same way - they're just strings used for validation. The code doesn't have any model-specific logic. Documentation listing specific models would get outdated quickly.
  The comment should be deleted as it asks for unnecessary work and violates the rule about asking authors to verify/ensure things. The model string additions don't require special testing or documentation updates.

2. extensions/llms/bedrock/pandasai_bedrock/claude.py:37

• Draft comment:
  New models added to _supported__models list. Please ensure corresponding unit tests are added to verify cross-region inference functionality.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50%
  The comment is about testing new functionality - regional model variants. However, looking at the code, there's no special handling for regional models - they're just strings in a list. The actual region handling would be done by the bedrock_runtime_client which is passed in. The model strings are only used for validation in init. Adding unit tests just for new model string constants seems unnecessary.
  The comment raises a valid point about testing new functionality. Regional inference could have important implications for production use.
  While regional testing is important, it would be handled at the AWS SDK/client level, not in this code. This change is just adding new valid model string constants.
  The comment should be deleted as it requests tests for functionality that isn't actually implemented in this code - the regional handling is done by AWS's client.

Workflow ID: wflow_v0qO67AOmIQeFWOq

---

You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.

[ArslanSaleem]

@brunobpr PandasAI already supports LiteLLM, which allows you to achieve this functionality with any LLM provider. We recommend using that approach to keep things consistent and flexible.

https://docs.pandas-ai.com/v3/large-language-models#how-to-set-up-any-llm%3F