Fix vLLM prompt logprobs calculation #384

pan-x-c · 2025-11-13T11:09:56Z

Description

vLLM does not apply temperature scaling to logits when calculating prompt logprobs.
This PR adds the missing temperature scaling with a patch.

Checklist

Please check the following items before code is ready to be reviewed.

Code has passed all tests
Docstrings have been added/updated in Google Style
Documentation has been updated
Code is ready for review

gemini-code-assist · 2025-11-13T11:10:17Z

Summary of Changes

Hello @pan-x-c, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request addresses an issue with the calculation of prompt log probabilities in vLLM by introducing a targeted patch. It also involves a refactoring of the existing vLLM patching infrastructure, centralizing version detection and organizing API-related patches. The changes ensure accurate logprob computation, especially when temperature sampling is involved, and update the minimum required vLLM version to accommodate these fixes.

Highlights

vLLM Dependency Update: The minimum required vllm version has been updated from 0.9.1 to 0.10.0 in pyproject.toml to align with the introduced fixes and features.
vLLM Patching Refactor: The vllm_patch utilities have been reorganized into a new trinity.common.models.vllm_patch package. This includes moving the get_vllm_version function and renaming api/vllm_patch.py to vllm_patch/api_patch.py for better modularity.
Prompt Logprobs Calculation Fix: A new patch (patch_vllm_prompt_logprobs) has been introduced to correct the calculation of prompt log probabilities within the vLLM GPUModelRunner. This fix specifically ensures that temperature scaling is correctly applied to logits before logprob computation, addressing an accuracy issue.
Integration of New Patch: The newly implemented prompt logprobs patch is now applied during the initialization of the VLLMWorker, ensuring that all vLLM operations benefit from the corrected logprob calculation.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	`/gemini review`	Performs a code review for the current pull request in its current state.
Pull Request Summary	`/gemini summary`	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	`/gemini help`	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

gemini-code-assist

Code Review

This pull request correctly fixes an issue with prompt logprob calculation in vLLM by applying temperature scaling, which was missing in the original implementation. The refactoring of vLLM patches into a dedicated directory is a good move for better code organization. My review includes a high-severity fix for a potential bug in the temperature handling logic and a medium-severity suggestion to improve the maintainability of the monkey-patched function by adding documentation about its origin and purpose.

trinity/common/models/vllm_patch/worker_patch.py

pan-x-c · 2025-11-13T11:18:13Z

/unittest-module-common

github-actions · 2025-11-13T11:24:44Z

Summary

Tests 📝	Passed ✅	Failed ❌	Skipped ⏭️	Other ❓	Flaky 🍂	Duration ⏱️
34	34	0	0	0	0	4m 57s

Tests

Test Name	Status	Duration
tests/common/config_test.py::TestConfig::test_all_examples_are_valid	✅	41.8s
tests/common/config_test.py::TestConfig::test_config_flatten	✅	39ms
tests/common/config_test.py::TestConfig::test_continue_from_checkpoint_is_valid	✅	188ms
tests/common/config_test.py::TestConfig::test_default_workflow	✅	89ms
tests/common/config_test.py::TestConfig::test_load_default_config	✅	3.2s
tests/common/config_test.py::TestConfig::test_max_token_len_per_gpu_set_correctly	✅	91ms
tests/common/config_test.py::TestConfig::test_optimizer_config_propagation	✅	89ms
tests/common/config_test.py::TestConfig::test_update_config_from_ray_cluster	✅	1.7s
tests/common/experience_test.py::TestEID::test_eid_properties	✅	1ms
tests/common/experience_test.py::TestExperience::test_action_mask_and_logprobs_type	✅	1ms
tests/common/experience_test.py::TestExperience::test_assertions	✅	1ms
tests/common/experience_test.py::TestExperience::test_dpo_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_gather	✅	1ms
tests/common/experience_test.py::TestExperience::test_hf_datasets_conversion	✅	15ms
tests/common/experience_test.py::TestExperience::test_multi_turn_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_serialize_deserialize	✅	2ms
tests/common/experience_test.py::TestExperience::test_single_turn_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_to_dict	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_batch_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_dpo_experience_batch_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_experience_model_experience_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_gather_experiences_with_custom_fields	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_multiturn_experience_batch_converstion	✅	1ms
tests/common/vllm_test.py::ModelWrapperTest_0::test_generate	✅	53.2s
tests/common/vllm_test.py::ModelWrapperTest_1::test_generate	✅	32.6s
tests/common/vllm_test.py::ModelWrapperTest_2::test_generate	✅	43.1s
tests/common/vllm_test.py::TestModelLen_0::test_model_len	✅	17.4s
tests/common/vllm_test.py::TestModelLen_1::test_model_len	✅	16.9s
tests/common/vllm_test.py::TestAPIServer::test_api	✅	22.6s
tests/common/vllm_test.py::TestAsyncAPIServer::test_api_async	✅	22.8s
tests/common/vllm_test.py::TestTokenizer::test_action_mask	✅	288ms
tests/common/vllm_test.py::TestTokenizer::test_action_mask_with_tools	✅	253ms
tests/common/vllm_test.py::TestAPIServerToolCall_0_deepseek_r1::test_api_tool_calls	✅	19.7s
tests/common/vllm_test.py::TestAPIServerToolCall_1::test_api_tool_calls	✅	17.4s

Github Test Reporter by CTRF 💚

pan-x-c · 2025-11-13T13:50:53Z

/unittest-module-common

github-actions · 2025-11-13T13:57:34Z

Summary

Tests 📝	Passed ✅	Failed ❌	Skipped ⏭️	Other ❓	Flaky 🍂	Duration ⏱️
35	32	3	0	0	0	5m 6s

Failed Tests

Failed Tests ❌	Fail Message
❌ tests/common/vllm_test.py::TestAPIServer::test_api	The test failed in the call phase due to an assertion error
❌ tests/common/vllm_test.py::TestLogprobs::test_logprobs	The test failed in the call phase due to an assertion error
❌ tests/common/vllm_test.py::TestAsyncAPIServer::test_api_async	The test failed in the call phase due to an assertion error

Tests

Test Name	Status	Duration
tests/common/config_test.py::TestConfig::test_all_examples_are_valid	✅	41.3s
tests/common/config_test.py::TestConfig::test_config_flatten	✅	40ms
tests/common/config_test.py::TestConfig::test_continue_from_checkpoint_is_valid	✅	445ms
tests/common/config_test.py::TestConfig::test_default_workflow	✅	89ms
tests/common/config_test.py::TestConfig::test_load_default_config	✅	3.6s
tests/common/config_test.py::TestConfig::test_max_token_len_per_gpu_set_correctly	✅	90ms
tests/common/config_test.py::TestConfig::test_optimizer_config_propagation	✅	90ms
tests/common/config_test.py::TestConfig::test_update_config_from_ray_cluster	✅	1.8s
tests/common/experience_test.py::TestEID::test_eid_properties	✅	1ms
tests/common/experience_test.py::TestExperience::test_action_mask_and_logprobs_type	✅	1ms
tests/common/experience_test.py::TestExperience::test_assertions	✅	1ms
tests/common/experience_test.py::TestExperience::test_dpo_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_gather	✅	1ms
tests/common/experience_test.py::TestExperience::test_hf_datasets_conversion	✅	15ms
tests/common/experience_test.py::TestExperience::test_multi_turn_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_serialize_deserialize	✅	1ms
tests/common/experience_test.py::TestExperience::test_single_turn_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_to_dict	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_batch_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_dpo_experience_batch_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_experience_model_experience_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_gather_experiences_with_custom_fields	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_multiturn_experience_batch_converstion	✅	1ms
tests/common/vllm_test.py::ModelWrapperTest_0::test_generate	✅	51.4s
tests/common/vllm_test.py::ModelWrapperTest_1::test_generate	✅	32.2s
tests/common/vllm_test.py::ModelWrapperTest_2::test_generate	✅	42.3s
tests/common/vllm_test.py::TestModelLen_0::test_model_len	✅	16.9s
tests/common/vllm_test.py::TestModelLen_1::test_model_len	✅	17.0s
tests/common/vllm_test.py::TestAPIServer::test_api	❌	18.9s
tests/common/vllm_test.py::TestLogprobs::test_logprobs	❌	18.8s
tests/common/vllm_test.py::TestAsyncAPIServer::test_api_async	❌	19.0s
tests/common/vllm_test.py::TestTokenizer::test_action_mask	✅	246ms
tests/common/vllm_test.py::TestTokenizer::test_action_mask_with_tools	✅	236ms
tests/common/vllm_test.py::TestAPIServerToolCall_0_deepseek_r1::test_api_tool_calls	✅	19.5s
tests/common/vllm_test.py::TestAPIServerToolCall_1::test_api_tool_calls	✅	18.0s

Github Test Reporter by CTRF 💚

tests/common/vllm_test.py

pan-x-c · 2025-11-14T04:26:48Z

/unittest-module-common

github-actions · 2025-11-14T04:33:33Z

Summary

Tests 📝	Passed ✅	Failed ❌	Skipped ⏭️	Other ❓	Flaky 🍂	Duration ⏱️
35	34	1	0	0	0	5m 13s

Failed Tests

Failed Tests ❌	Fail Message
❌ tests/common/vllm_test.py::TestLogprobs::test_logprobs	The test failed in the call phase due to an assertion error

Tests

Test Name	Status	Duration
tests/common/config_test.py::TestConfig::test_all_examples_are_valid	✅	41.2s
tests/common/config_test.py::TestConfig::test_config_flatten	✅	39ms
tests/common/config_test.py::TestConfig::test_continue_from_checkpoint_is_valid	✅	442ms
tests/common/config_test.py::TestConfig::test_default_workflow	✅	88ms
tests/common/config_test.py::TestConfig::test_load_default_config	✅	4.1s
tests/common/config_test.py::TestConfig::test_max_token_len_per_gpu_set_correctly	✅	89ms
tests/common/config_test.py::TestConfig::test_optimizer_config_propagation	✅	88ms
tests/common/config_test.py::TestConfig::test_update_config_from_ray_cluster	✅	1.8s
tests/common/experience_test.py::TestEID::test_eid_properties	✅	1ms
tests/common/experience_test.py::TestExperience::test_action_mask_and_logprobs_type	✅	1ms
tests/common/experience_test.py::TestExperience::test_assertions	✅	1ms
tests/common/experience_test.py::TestExperience::test_dpo_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_gather	✅	1ms
tests/common/experience_test.py::TestExperience::test_hf_datasets_conversion	✅	15ms
tests/common/experience_test.py::TestExperience::test_multi_turn_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_serialize_deserialize	✅	1ms
tests/common/experience_test.py::TestExperience::test_single_turn_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_to_dict	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_batch_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_dpo_experience_batch_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_experience_model_experience_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_gather_experiences_with_custom_fields	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_multiturn_experience_batch_converstion	✅	1ms
tests/common/vllm_test.py::ModelWrapperTest_0::test_generate	✅	52.0s
tests/common/vllm_test.py::ModelWrapperTest_1::test_generate	✅	31.5s
tests/common/vllm_test.py::ModelWrapperTest_2::test_generate	✅	42.5s
tests/common/vllm_test.py::TestModelLen_0::test_model_len	✅	17.9s
tests/common/vllm_test.py::TestModelLen_1::test_model_len	✅	16.9s
tests/common/vllm_test.py::TestAPIServer::test_api	✅	22.3s
tests/common/vllm_test.py::TestLogprobs::test_logprobs	❌	18.2s
tests/common/vllm_test.py::TestAsyncAPIServer::test_api_async	✅	22.5s
tests/common/vllm_test.py::TestTokenizer::test_action_mask	✅	255ms
tests/common/vllm_test.py::TestTokenizer::test_action_mask_with_tools	✅	251ms
tests/common/vllm_test.py::TestAPIServerToolCall_0_deepseek_r1::test_api_tool_calls	✅	19.4s
tests/common/vllm_test.py::TestAPIServerToolCall_1::test_api_tool_calls	✅	18.0s

Github Test Reporter by CTRF 💚

pan-x-c · 2025-11-14T05:58:16Z

/unittest-module-common

github-actions · 2025-11-14T06:05:05Z

Summary

Tests 📝	Passed ✅	Failed ❌	Skipped ⏭️	Other ❓	Flaky 🍂	Duration ⏱️
35	35	0	0	0	0	5m 16s

Tests

Test Name	Status	Duration
tests/common/config_test.py::TestConfig::test_all_examples_are_valid	✅	41.4s
tests/common/config_test.py::TestConfig::test_config_flatten	✅	39ms
tests/common/config_test.py::TestConfig::test_continue_from_checkpoint_is_valid	✅	443ms
tests/common/config_test.py::TestConfig::test_default_workflow	✅	89ms
tests/common/config_test.py::TestConfig::test_load_default_config	✅	3.5s
tests/common/config_test.py::TestConfig::test_max_token_len_per_gpu_set_correctly	✅	89ms
tests/common/config_test.py::TestConfig::test_optimizer_config_propagation	✅	91ms
tests/common/config_test.py::TestConfig::test_update_config_from_ray_cluster	✅	1.8s
tests/common/experience_test.py::TestEID::test_eid_properties	✅	1ms
tests/common/experience_test.py::TestExperience::test_action_mask_and_logprobs_type	✅	1ms
tests/common/experience_test.py::TestExperience::test_assertions	✅	1ms
tests/common/experience_test.py::TestExperience::test_dpo_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_gather	✅	1ms
tests/common/experience_test.py::TestExperience::test_hf_datasets_conversion	✅	15ms
tests/common/experience_test.py::TestExperience::test_multi_turn_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_serialize_deserialize	✅	2ms
tests/common/experience_test.py::TestExperience::test_single_turn_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_to_dict	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_batch_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_dpo_experience_batch_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_experience_model_experience_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_gather_experiences_with_custom_fields	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_multiturn_experience_batch_converstion	✅	1ms
tests/common/vllm_test.py::ModelWrapperTest_0::test_generate	✅	53.0s
tests/common/vllm_test.py::ModelWrapperTest_1::test_generate	✅	31.6s
tests/common/vllm_test.py::ModelWrapperTest_2::test_generate	✅	42.3s
tests/common/vllm_test.py::TestModelLen_0::test_model_len	✅	17.7s
tests/common/vllm_test.py::TestModelLen_1::test_model_len	✅	17.0s
tests/common/vllm_test.py::TestAPIServer::test_api	✅	22.4s
tests/common/vllm_test.py::TestLogprobs::test_logprobs	✅	19.1s
tests/common/vllm_test.py::TestAsyncAPIServer::test_api_async	✅	23.4s
tests/common/vllm_test.py::TestTokenizer::test_action_mask	✅	228ms
tests/common/vllm_test.py::TestTokenizer::test_action_mask_with_tools	✅	224ms
tests/common/vllm_test.py::TestAPIServerToolCall_0_deepseek_r1::test_api_tool_calls	✅	19.5s
tests/common/vllm_test.py::TestAPIServerToolCall_1::test_api_tool_calls	✅	18.0s

Github Test Reporter by CTRF 💚

hiyuchang · 2025-11-14T06:32:23Z

/unittest-all

github-actions · 2025-11-14T07:48:13Z

Summary

Tests 📝	Passed ✅	Failed ❌	Skipped ⏭️	Other ❓	Flaky 🍂	Duration ⏱️
174	171	0	3	0	0	1h 14m

Skipped

Tests	Status
tests/explorer/workflow_test.py::TestAgentScopeWorkflowAdapter::test_adapter	skipped ⏭️
tests/trainer/trainer_test.py::TestMultiModalGRPO::test_trainer	skipped ⏭️
tests/trainer/trainer_test.py::TestMultiModalSFT::test_trainer	skipped ⏭️

Tests

Test Name	Status	Duration
tests/algorithm/advantage_fn_test.py::TestGroupedAdvantageFn::test_batch_level_std_grpo	✅	41ms
tests/algorithm/advantage_fn_test.py::TestGroupedAdvantageFn::test_batch_level_step_wise_grpo_advantage	✅	2ms
tests/algorithm/advantage_fn_test.py::TestGroupedAdvantageFn::test_duplicate_grpo	✅	5ms
tests/algorithm/advantage_fn_test.py::TestGroupedAdvantageFn::test_grpo_advantage	✅	3ms
tests/algorithm/advantage_fn_test.py::TestGroupedAdvantageFn::test_grpo_correct_bias	✅	2ms
tests/algorithm/advantage_fn_test.py::TestGroupedAdvantageFn::test_grpo_reward_std	✅	1ms
tests/algorithm/advantage_fn_test.py::TestGroupedAdvantageFn::test_step_wise_grpo_advantage	✅	2ms
tests/algorithm/advantage_fn_test.py::TestGroupedAdvantageFn::test_step_wise_grpo_with_std_threshold	✅	2ms
tests/algorithm/policy_loss_test.py::VerlPolicyLossTest::test_dpo_policy_loss	✅	2ms
tests/algorithm/policy_loss_test.py::VerlPolicyLossTest::test_gspo_policy_loss	✅	1ms
tests/algorithm/policy_loss_test.py::VerlPolicyLossTest::test_mix_policy_loss	✅	1ms
tests/algorithm/policy_loss_test.py::VerlPolicyLossTest::test_opmd_policy_loss	✅	1ms
tests/algorithm/policy_loss_test.py::VerlPolicyLossTest::test_ppo_policy_loss	✅	1ms
tests/algorithm/policy_loss_test.py::VerlPolicyLossTest::test_sft_policy_loss	✅	1ms
tests/buffer/experience_pipeline_test.py::TestExperiencePipeline::test_experience_pipeline	✅	19.5s
tests/buffer/experience_pipeline_test.py::TestExperiencePipeline::test_pass_rate_calculation	✅	15.4s
tests/buffer/experience_storage_test.py::ExperienceStorageTest::test_sql_experience_buffer	✅	4.0s
tests/buffer/experience_storage_test.py::ExperienceStorageTest::test_sql_storage_0_sft	✅	5.9s
tests/buffer/experience_storage_test.py::ExperienceStorageTest::test_sql_storage_1_dpo	✅	6.5s
tests/buffer/file_test.py::TestFileBuffer::test_file_reader	✅	154ms
tests/buffer/file_test.py::TestFileBuffer::test_file_writer	✅	4.4s
tests/buffer/formatter_test.py::TestFormatter::test_dpo_messages_formatter	✅	528ms
tests/buffer/formatter_test.py::TestFormatter::test_dpo_plaintext_formatter	✅	493ms
tests/buffer/formatter_test.py::TestFormatter::test_multi_modal_sft_formatter	✅	811ms
tests/buffer/formatter_test.py::TestFormatter::test_sft_messages_formatter	✅	976ms
tests/buffer/formatter_test.py::TestFormatter::test_sft_plaintext_formatter	✅	723ms
tests/buffer/formatter_test.py::TestFormatter::test_task_formatter	✅	223ms
tests/buffer/queue_test.py::TestQueueBuffer::test_priority_queue_buffer_reuse	✅	8.7s
tests/buffer/queue_test.py::TestQueueBuffer::test_priority_queue_capacity	✅	4.9s
tests/buffer/queue_test.py::TestQueueBuffer::test_priority_queue_reuse_count_control	✅	6.7s
tests/buffer/queue_test.py::TestQueueBuffer::test_queue_buffer_0_queue	✅	5.9s
tests/buffer/queue_test.py::TestQueueBuffer::test_queue_buffer_1_priority_queue	✅	5.8s
tests/buffer/queue_test.py::TestQueueBuffer::test_queue_buffer_capacity	✅	6.5s
tests/buffer/reward_shaping_mapper_test.py::TestRewardShapingMapper::test_basic_usage	✅	6ms
tests/buffer/sql_test.py::TestSQLBuffer::test_sql_buffer_read_write	✅	4.5s
tests/buffer/task_scheduler_test.py::TestTaskScheduler::test_task_scheduler_0	✅	177ms
tests/buffer/task_scheduler_test.py::TestTaskScheduler::test_task_scheduler_1	✅	157ms
tests/buffer/task_scheduler_test.py::TestTaskScheduler::test_task_scheduler_2	✅	160ms
tests/buffer/task_scheduler_test.py::TestTaskScheduler::test_task_scheduler_3	✅	162ms
tests/buffer/task_scheduler_test.py::TestTaskScheduler::test_task_scheduler_4	✅	177ms
tests/buffer/task_storage_test.py::TaskStorageTest::test_read_task_0	✅	68ms
tests/buffer/task_storage_test.py::TaskStorageTest::test_read_task_1	✅	4.2s
tests/buffer/task_storage_test.py::TaskStorageTest::test_read_task_2	✅	47ms
tests/buffer/task_storage_test.py::TaskStorageTest::test_read_task_3	✅	3.9s
tests/buffer/task_storage_test.py::TaskStorageTest::test_read_task_4	✅	48ms
tests/buffer/task_storage_test.py::TaskStorageTest::test_read_task_5	✅	4.3s
tests/cli/launcher_test.py::TestLauncherMain::test_debug_mode	✅	45.9s
tests/cli/launcher_test.py::TestLauncherMain::test_main_run_command	✅	6.6s
tests/cli/launcher_test.py::TestLauncherMain::test_main_run_in_dlc	✅	1.5s
tests/cli/launcher_test.py::TestLauncherMain::test_main_studio_command	✅	320ms
tests/cli/launcher_test.py::TestLauncherMain::test_multi_stage_run	✅	1.7s
tests/common/config_test.py::TestConfig::test_all_examples_are_valid	✅	33.1s
tests/common/config_test.py::TestConfig::test_config_flatten	✅	39ms
tests/common/config_test.py::TestConfig::test_continue_from_checkpoint_is_valid	✅	186ms
tests/common/config_test.py::TestConfig::test_default_workflow	✅	88ms
tests/common/config_test.py::TestConfig::test_load_default_config	✅	3.4s
tests/common/config_test.py::TestConfig::test_max_token_len_per_gpu_set_correctly	✅	89ms
tests/common/config_test.py::TestConfig::test_optimizer_config_propagation	✅	88ms
tests/common/config_test.py::TestConfig::test_update_config_from_ray_cluster	✅	155ms
tests/common/experience_test.py::TestEID::test_eid_properties	✅	1ms
tests/common/experience_test.py::TestExperience::test_action_mask_and_logprobs_type	✅	1ms
tests/common/experience_test.py::TestExperience::test_assertions	✅	1ms
tests/common/experience_test.py::TestExperience::test_dpo_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_gather	✅	1ms
tests/common/experience_test.py::TestExperience::test_hf_datasets_conversion	✅	15ms
tests/common/experience_test.py::TestExperience::test_multi_turn_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_serialize_deserialize	✅	1ms
tests/common/experience_test.py::TestExperience::test_single_turn_experience	✅	1ms
tests/common/experience_test.py::TestExperience::test_to_dict	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_batch_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_dpo_experience_batch_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_experience_model_experience_conversion	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_gather_experiences_with_custom_fields	✅	1ms
tests/common/experience_test.py::TestExperienceConversion::test_multiturn_experience_batch_converstion	✅	1ms
tests/common/vllm_test.py::ModelWrapperTest_0::test_generate	✅	44.9s
tests/common/vllm_test.py::ModelWrapperTest_1::test_generate	✅	32.4s
tests/common/vllm_test.py::ModelWrapperTest_2::test_generate	✅	43.8s
tests/common/vllm_test.py::TestModelLen_0::test_model_len	✅	17.6s
tests/common/vllm_test.py::TestModelLen_1::test_model_len	✅	16.9s
tests/common/vllm_test.py::TestAPIServer::test_api	✅	22.6s
tests/common/vllm_test.py::TestLogprobs::test_logprobs	✅	18.8s
tests/common/vllm_test.py::TestAsyncAPIServer::test_api_async	✅	22.7s
tests/common/vllm_test.py::TestTokenizer::test_action_mask	✅	252ms
tests/common/vllm_test.py::TestTokenizer::test_action_mask_with_tools	✅	227ms
tests/common/vllm_test.py::TestAPIServerToolCall_0_deepseek_r1::test_api_tool_calls	✅	20.0s
tests/common/vllm_test.py::TestAPIServerToolCall_1::test_api_tool_calls	✅	17.5s
tests/explorer/explorer_test.py::TestExplorerCountdownEval::test_explorer	✅	1m 9s
tests/explorer/explorer_test.py::TestExplorerGSM8KRULERNoEval::test_explorer	✅	1m 39s
tests/explorer/explorer_test.py::TestExplorerGSM8k::test_explorer	✅	3m 37s
tests/explorer/explorer_test.py::ServeTest::test_serve	✅	1m 19s
tests/explorer/scheduler_test.py::SchedulerTest::test_async_workflow	✅	12.9s
tests/explorer/scheduler_test.py::SchedulerTest::test_concurrent_operations	✅	12.3s
tests/explorer/scheduler_test.py::SchedulerTest::test_get_results	✅	30.5s
tests/explorer/scheduler_test.py::SchedulerTest::test_multi_step_execution	✅	12.6s
tests/explorer/scheduler_test.py::SchedulerTest::test_non_repeatable_workflow	✅	12.4s
tests/explorer/scheduler_test.py::SchedulerTest::test_scheduler_all_methods	✅	22.2s
tests/explorer/scheduler_test.py::SchedulerTest::test_scheduler_restart_after_stop	✅	24.0s
tests/explorer/scheduler_test.py::SchedulerTest::test_split_tasks	✅	15.5s
tests/explorer/scheduler_test.py::SchedulerTest::test_stepwise_experience_eid	✅	13.0s
tests/explorer/scheduler_test.py::SchedulerTest::test_wait_all	✅	15.5s
tests/explorer/scheduler_test.py::SchedulerTest::test_wait_all_timeout_with_multi_batch	✅	20.9s
tests/explorer/step_wise_workflow_test.py::WorkflowTest::test_reward_propagation_workflow_0	✅	2ms
tests/explorer/step_wise_workflow_test.py::WorkflowTest::test_reward_propagation_workflow_1	✅	602ms
tests/explorer/step_wise_workflow_test.py::WorkflowTest::test_step_wise_reward_workflow_0	✅	1ms
tests/explorer/step_wise_workflow_test.py::WorkflowTest::test_step_wise_reward_workflow_1	✅	1.0s
tests/explorer/step_wise_workflow_test.py::WorkflowTest::test_workflows_raise_error	✅	1ms
tests/explorer/step_wise_workflow_test.py::WorkflowTest::test_workflows_stop_at_max_env_steps	✅	1.0s
tests/explorer/workflow_test.py::WorkflowTest::test_gsm8k_workflow	✅	15ms
tests/explorer/workflow_test.py::WorkflowTest::test_math_boxed_workflow	✅	24ms
tests/explorer/workflow_test.py::WorkflowTest::test_math_complex_workflow	✅	268ms
tests/explorer/workflow_test.py::WorkflowTest::test_math_eval_workflow	✅	4ms
tests/explorer/workflow_test.py::WorkflowTest::test_math_fraction_workflow	✅	17ms
tests/explorer/workflow_test.py::WorkflowTest::test_math_workflow	✅	10ms
tests/explorer/workflow_test.py::WorkflowTest::test_rm_gallery_workflow	✅	79ms
tests/explorer/workflow_test.py::WorkflowTest::test_workflow_repeatable_0	✅	1ms
tests/explorer/workflow_test.py::WorkflowTest::test_workflow_repeatable_1	✅	101ms
tests/explorer/workflow_test.py::WorkflowTest::test_workflow_resettable_0	✅	1ms
tests/explorer/workflow_test.py::WorkflowTest::test_workflow_resettable_1	✅	201ms
tests/explorer/workflow_test.py::MultiTurnWorkflowTest_0::test_multi_turn_workflow	✅	14.4s
tests/explorer/workflow_test.py::MultiTurnWorkflowTest_1::test_multi_turn_workflow	✅	14.7s
tests/explorer/workflow_test.py::TestAgentScopeWorkflowAdapter::test_adapter	⏭️	1ms
tests/explorer/workflow_test.py::TestWorkflowRunner::test_workflow_runner	✅	296ms
tests/manager/synchronizer_test.py::TestSynchronizerExit::test_synchronizer	✅	1m
tests/manager/synchronizer_test.py::TestStateDictBasedSynchronizer_0::test_synchronizer	✅	1m 44s
tests/manager/synchronizer_test.py::TestStateDictBasedSynchronizer_1::test_synchronizer	✅	1m 46s
tests/manager/synchronizer_test.py::TestStateDictBasedSynchronizer_2::test_synchronizer	✅	2m 27s
tests/manager/synchronizer_test.py::TestStateDictBasedSynchronizer_3::test_synchronizer	✅	2m 39s
tests/manager/synchronizer_test.py::TestNCCLBasedSynchronizer_0::test_synchronizer	✅	1m 44s
tests/manager/synchronizer_test.py::TestNCCLBasedSynchronizer_1::test_synchronizer	✅	1m 44s
tests/service/data_juicer_test.py::TestDataJuicer::test_config	✅	1.3s
tests/service/data_juicer_test.py::TestDataJuicer::test_server_start	✅	21.6s
tests/service/data_juicer_test.py::TestDataJuicerExperiencePipeline::test_data_juicer_operators	✅	30.8s
tests/service/data_juicer_test.py::TestDataJuicerTaskPipeline::test_data_juicer_task_pipeline	✅	14.4s
tests/trainer/trainer_test.py::TestTrainerCountdown_0_fsdp::test_trainer	✅	2m 52s
tests/trainer/trainer_test.py::TestTrainerCountdown_1_megatron::test_trainer	✅	5m
tests/trainer/trainer_test.py::TestStepAheadAsyncRL::test_trainer	✅	1m 16s
tests/trainer/trainer_test.py::TestTrainerGSM8K_0_fsdp::test_trainer	✅	1m 20s
tests/trainer/trainer_test.py::TestTrainerGSM8K_1_fsdp2::test_trainer	✅	1m 19s
tests/trainer/trainer_test.py::TestTrainerGSM8K_2_fsdp::test_trainer	✅	1m 20s
tests/trainer/trainer_test.py::TestTrainerGSM8K_3_fsdp2::test_trainer	✅	1m 34s
tests/trainer/trainer_test.py::TestTrainerSFTWarmupGSM8K::test_trainer	✅	2m 27s
tests/trainer/trainer_test.py::TestTrainerDPO::test_trainer	✅	1m 1s
tests/trainer/trainer_test.py::TestTrainerSFT::test_trainer	✅	57.4s
tests/trainer/trainer_test.py::TestTrainerToolsSFT::test_trainer_tools	✅	58.0s
tests/trainer/trainer_test.py::TestFullyAsyncMode_0_fsdp::test_fully_async_mode	✅	1m 49s
tests/trainer/trainer_test.py::TestFullyAsyncMode_1_fsdp::test_fully_async_mode	✅	1m 49s
tests/trainer/trainer_test.py::TestFullyAsyncMode_2_megatron::test_fully_async_mode	✅	2m 55s
tests/trainer/trainer_test.py::TestTrainerCheckpointSave_0_fsdp::test_trainer	✅	2m 19s
tests/trainer/trainer_test.py::TestTrainerCheckpointSave_1_megatron::test_trainer	✅	5m 33s
tests/trainer/trainer_test.py::TestTrainerMIX::test_trainer	✅	1m 22s
tests/trainer/trainer_test.py::TestMultiModalGRPO::test_trainer	⏭️	810ms
tests/trainer/trainer_test.py::TestMultiModalSFT::test_trainer	⏭️	809ms
tests/trainer/trainer_test.py::TestTrainerLoRA::test_trainer	✅	3m 9s
tests/utils/eval_utils_test.py::TestComputeScore::test_both_boxed_and_equivalent	✅	15ms
tests/utils/eval_utils_test.py::TestComputeScore::test_both_boxed_and_not_equivalent	✅	1ms
tests/utils/eval_utils_test.py::TestComputeScore::test_empty_ground_truth	✅	2ms
tests/utils/eval_utils_test.py::TestComputeScore::test_empty_solution_string	✅	1ms
tests/utils/eval_utils_test.py::TestComputeScore::test_multiple_boxed_answers_in_solution	✅	2ms
tests/utils/eval_utils_test.py::TestComputeScore::test_solution_boxed_truth_raw_and_equivalent	✅	1ms
tests/utils/eval_utils_test.py::TestComputeScore::test_solution_boxed_truth_raw_and_not_equivalent	✅	1ms
tests/utils/eval_utils_test.py::TestComputeScore::test_solution_not_boxed	✅	1ms
tests/utils/eval_utils_test.py::TestComputeScore::test_solution_raw_and_ground_truth_boxed_equivalent	✅	1ms
tests/utils/eval_utils_test.py::TestMathEvalUtils::test_extract_answer	✅	4ms
tests/utils/eval_utils_test.py::TestMathEvalUtils::test_verify_math_answer	✅	73ms
tests/utils/eval_utils_test.py::TestEvalUtils::test_is_equiv	✅	6ms
tests/utils/log_test.py::LogTest::test_actor_log	✅	5.1s
tests/utils/log_test.py::LogTest::test_group_by_node	✅	4.8s
tests/utils/log_test.py::LogTest::test_no_actor_log	✅	899ms
tests/utils/plugin_test.py::TestPluginLoader::test_load_plugins_local_0__workspace_tests_utils_plugins	✅	90ms
tests/utils/plugin_test.py::TestPluginLoader::test_load_plugins_local_1_tests_utils_plugins	✅	89ms
tests/utils/plugin_test.py::TestPluginLoader::test_load_plugins_remote_0__workspace_tests_utils_plugins	✅	22.3s
tests/utils/plugin_test.py::TestPluginLoader::test_load_plugins_remote_1_tests_utils_plugins	✅	22.1s
tests/utils/plugin_test.py::TestPluginLoader::test_passing_custom_class_0__workspace_tests_utils_plugins	✅	11.8s
tests/utils/plugin_test.py::TestPluginLoader::test_passing_custom_class_1_tests_utils_plugins	✅	11.7s

Github Test Reporter by CTRF 💚

pan-x-c added 2 commits November 13, 2025 18:40

patch vllm

cd0173b

update patch

b4f0977

gemini-code-assist bot reviewed Nov 13, 2025

View reviewed changes

trinity/common/models/vllm_patch/worker_patch.py Outdated Show resolved Hide resolved

trinity/common/models/vllm_patch/worker_patch.py Show resolved Hide resolved

update patch

7dc051a

limit vllm version

8abbf0b

pan-x-c added 2 commits November 13, 2025 19:41

add temperature in logprobs/convert interface

cff84bf

add vllm tests

97fce66

hiyuchang reviewed Nov 14, 2025

View reviewed changes

tests/common/vllm_test.py Outdated Show resolved Hide resolved

pan-x-c added 2 commits November 14, 2025 12:07

update commits

1a64af8

fix logprobs tests

e7b0345

fix vllm logprobs

1ddb836

update sphinx timeout

9164c86

hiyuchang approved these changes Nov 14, 2025

View reviewed changes

hiyuchang merged commit 81b3554 into modelscope:main Nov 14, 2025
1 check passed

Fix vLLM prompt logprobs calculation #384

Fix vLLM prompt logprobs calculation #384

Uh oh!

Conversation

pan-x-c commented Nov 13, 2025 • edited by hiyuchang Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Checklist

Uh oh!

gemini-code-assist bot commented Nov 13, 2025

Summary of Changes

Highlights

Footnotes

Uh oh!

gemini-code-assist bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

Uh oh!

Uh oh!

pan-x-c commented Nov 13, 2025

Uh oh!

github-actions bot commented Nov 13, 2025

Summary

Tests

Uh oh!

pan-x-c commented Nov 13, 2025

Uh oh!

github-actions bot commented Nov 13, 2025

Summary

Failed Tests

Tests

Uh oh!

Uh oh!

pan-x-c commented Nov 14, 2025

Uh oh!

github-actions bot commented Nov 14, 2025

Summary

Failed Tests

Tests

Uh oh!

pan-x-c commented Nov 14, 2025

Uh oh!

github-actions bot commented Nov 14, 2025

Summary

Tests

Uh oh!

hiyuchang commented Nov 14, 2025

Uh oh!

github-actions bot commented Nov 14, 2025

Summary

Skipped

Tests

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

pan-x-c commented Nov 13, 2025 •

edited by hiyuchang

Loading