[llm] Add generate_from_pos API to LLM runner #11570
Merged
Conversation
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
As titled, this API allows us to support multi-turn conversation by passing in a `start_pos` argument to `generate_from_pos`.
larryliu0820
requested review from
iseeyuan,
jackzhxng and
swolchok
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/11570
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New FailureAs of commit 4644bb9 with merge base 72a095f ( NEW FAILURE - The following job has failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
Contributor
|
@larryliu0820 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
Contributor
|
@larryliu0820 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
mergennachin
approved these changes
Jun 13, 2025
larryliu0820
added a commit
that referenced
this pull request
Jun 17, 2025
This reverts commit be8ffd1.
larryliu0820
added a commit
that referenced
this pull request
Jun 17, 2025
Summary: This is a follow-up of #11570 (D76457271) We should not abort when num_prompt_tokens >= max_context_len, instead we should return error. Differential Revision: D76791781
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
As titled, this API allows us to support multi-turn conversation by passing in a
start_posargument togenerate_from_pos.This pull request introduces a new feature to support text generation from a specific starting position (
generate_from_pos) and includes updates to ensure proper error handling and functionality whenmax_new_tokensis negative. The changes primarily focus on extending theTextLLMRunnerclass and its associated methods to accommodate this new feature while maintaining backward compatibility.New Feature: Text Generation from a Specific Starting Position
Added
generate_from_posMethod: Introduced a new methodgenerate_from_posinTextLLMRunnerto allow text generation starting from a specified position in the KV cache. This includes updates to the method signature, logic, and error handling. (extension/llm/runner/text_llm_runner.cpp[1] [2] [3] [4];extension/llm/runner/text_llm_runner.h[5]Updated Documentation: Enhanced method documentation in
TextLLMRunnerto describe the new functionality, including parameters likestart_posand the expected behavior. (extension/llm/runner/text_llm_runner.h[1] [2]Error Handling Improvements
Validation for
max_new_tokens: Added checks to ensuremax_new_tokensis positive. If it is not, anInvalidArgumenterror is returned. This prevents invalid configurations during text generation. (extension/llm/runner/text_llm_runner.cppextension/llm/runner/text_llm_runner.cppR129-R156)Unit Test for Negative
max_new_tokens: Created a new test case (GenerateFromPosErrorsWithNegativeMaxNewTokens) to verify that thegenerate_from_posmethod correctly handles scenarios wheremax_new_tokensis negative. (extension/llm/runner/test/test_text_llm_runner.cppextension/llm/runner/test/test_text_llm_runner.cppR325-R379)