feat: Converse API Adapter: support upstream tokenization - #479
Open
AlexFoxalt wants to merge 2 commits into
Open
feat: Converse API Adapter: support upstream tokenization#479AlexFoxalt wants to merge 2 commits into
AlexFoxalt wants to merge 2 commits into
Conversation
adubovik
reviewed
Aug 12, 2026
| else None | ||
| ) | ||
| condition = configuration is not None and ( | ||
| "performanceConfig" in configuration |
adubovik
reviewed
Aug 12, 2026
| |Anthropic|Claude 5 Opus|anthropic.claude-opus-5|(text/image/document)-to-text|mantle: 🟢 \| legacy: 🟡|mantle: 🟢 \| legacy: 🟡|✅|✅|Anthropic SDK/Converse API| | ||
| |Anthropic|Claude 5 Sonnet|anthropic.claude-sonnet-5|(text/image/document)-to-text|mantle: 🟢 \| legacy: 🟡|mantle: 🟢 \| legacy: 🟡|✅|✅|Anthropic SDK/Converse API| | ||
| |Anthropic|Claude 5 Fable|anthropic.claude-fable-5|(text/image/document)-to-text|mantle: 🟢 \| legacy: 🟡|mantle: 🟢 \| legacy: 🟡|✅|✅|Anthropic SDK/Converse API| | ||
| |Anthropic|Claude 5 Fable|anthropic.claude-fable-5|(text/image/document)-to-text|mantle: 🟢 \| legacy: 🟢|mantle: 🟢 \| legacy: 🟢|✅|✅|Anthropic SDK/Converse API| |
Collaborator
There was a problem hiding this comment.
This doesn't look right - the legacy still doesn't support the precise tokenisation, because the PR didn't do anything with respect to the legacy client.
adubovik
reviewed
Aug 12, 2026
| _DEFAULT_CONVERSE_TOKENIZER_DEPLOYMENTS, | ||
| ids=lambda deployment: deployment.value, | ||
| ) | ||
| async def test_default_converse_tokenization_with_mixed_inputs( |
Collaborator
There was a problem hiding this comment.
The added tests are irrelevant to the PR.
In other words, they would have passed on the development too.
adubovik
reviewed
Aug 12, 2026
| request.original_request | ||
| ), | ||
| request=request if isinstance(request, Request) else None, | ||
| request=request |
Collaborator
There was a problem hiding this comment.
To make it work properly, we need to create an adapter for each Chat Completions request within TokenizeRequest, instead of reusing the same adapter for all sub-requests.
The same applies to the truncate_prompt request.
adubovik
requested changes
Aug 12, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
fixes #475
Summary
/tokenizeintegration coverage for mixedrequestandstringinputs across default and upstream tokenizer model groups.CountTokenssupport back to default Converse tokenization.