[TRTLLM-6352][feat] Migrate EAGLE3 and draft/target speculation to Drafter

[ziyixiong-nv]

Description

Main changes:

1. Update the class Drafter.
2. Create ModelDrafter for Drafter with draft model
3. Create Eagle3Drafter for SpeculativeDecodingMode.EAGLE3
4. Refactor the code in PyExecutor to use Drafter.

Test Coverage

tests/unittest/_torch/speculative/test_eagle3.py::test_llama_eagle3
tests/unittest/_torch/speculative/test_mtp.py
tests/unittest/_torch/speculative/test_draft_target.py

GitHub Bot Help

/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...

Provide a user friendly way for developers to interact with a Jenkins server.

Run /bot [-h|--help] to print this help message.

See details below for each supported subcommand.

run [--disable-fail-fast --skip-test --stage-list "A10-1, xxx" --gpu-type "A30, H100_PCIe" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-[Post-Merge]-1, xxx"]

Launch build/test pipelines. All previously running jobs will be killed.

--disable-fail-fast (OPTIONAL) : Disable fail fast on build/tests/infra failures.

--skip-test (OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.

--stage-list "A10-1, xxx" (OPTIONAL) : Only run the specified test stages. Examples: "A10-1, xxx". Note: Does NOT update GitHub check status.

--gpu-type "A30, H100_PCIe" (OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.

--only-multi-gpu-test (OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.

--disable-multi-gpu-test (OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.

--add-multi-gpu-test (OPTIONAL) : Force run the multi-GPU tests. Will also run L0 pre-merge pipeline.

--post-merge (OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.

--extra-stage "H100_PCIe-[Post-Merge]-1, xxx" (OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-[Post-Merge]-1, xxx".

For guidance on mapping tests to stage names, see docs/source/reference/ci-overview.md.

kill

kill

Kill all running builds associated with pull request.

skip

skip --comment COMMENT

Skip testing for latest commit on pull request. --comment "Reason for skipping build/test" is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

reuse-pipeline

reuse-pipeline

Reuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

Summary by CodeRabbit

• New Features

  • Introduced a new model-based drafter for generating draft tokens, supporting advanced speculative decoding modes.

• Improvements

  • Enhanced drafter interface to support resource management.
  • Updated drafter selection logic to support new modes and configurations.
  • Improved error handling and resource management in speculative token drafting.
  • Refined executor logic to integrate the new drafter approach, removing previous draft model engine handling.

• Bug Fixes

  • Corrected resource count and completion calculations in n-gram pool management.

• Documentation

  • Added and expanded docstrings for drafter classes and methods.

[ziyixiong-nv]

/bot run

[tensorrt-cicd]

PR_Github #11821 [ run ] triggered by Bot

[mikeiovine]

Thanks!

[tensorrt-cicd]

PR_Github #11821 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8761 completed with status: 'FAILURE'

[ziyixiong-nv]

/bot run

[ziyixiong-nv]

/bot run

[tensorrt-cicd]

PR_Github #11910 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #11911 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #11910 [ run ] completed with state ABORTED

[tensorrt-cicd]

PR_Github #11911 [ run ] completed with state FAILURE
/LLM/main/L0_MergeRequest_PR pipeline #8829 completed with status: 'FAILURE'

[ziyixiong-nv]

/bot run

[tensorrt-cicd]

PR_Github #11925 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #11925 [ run ] completed with state FAILURE
/LLM/main/L0_MergeRequest_PR pipeline #8843 completed with status: 'FAILURE'

[ziyixiong-nv]

/bot run

[ziyixiong-nv]

/bot reuse-pipeline

[tensorrt-cicd]

PR_Github #12205 [ reuse-pipeline ] triggered by Bot

[tensorrt-cicd]

PR_Github #12205 [ reuse-pipeline ] completed with state SUCCESS
Reusing PR_Github #12167 for commit df74a1a