[#7091][fix] remove INT8 from trtllm-bench --quantization choices

[Brumbelow]

[#7091][fix] Summary

• trtllm-bench build --help advertises INT8 under -q/--quantization, but passing it crashes in QuantConfig.from_quant_algo because QUANT_ALGO_LIST at tensorrt_llm/quantization/mode.py:52 deliberately excludes QuantAlgo.INT8 for weight quantization.
• Fixes [Bug]: INT8 quantization not supported despite being listed in help documentation #7091.

Description

• Drop INT8 from the VALID_QUANT_ALGOS Literal in tensorrt_llm/bench/utils/__init__.py.
• That Literal is the single source feeding both Click's quantization choices and the Pydantic field type, so removing it here makes both layers reject the value cleanly instead of letting it fall through to the AssertionError.

Test Coverage

• There is no existing tests/unittest/bench/ CLI harness, and the existing runtime assertion in QuantConfig.from_quant_algo continues to backstop the contract.
• trtllm-bench --model meta-llama/Llama-3.1-8B build --help no longer lists INT8 under --quantization (now 7 choices)
• --quantization INT8 is rejected by Click before any model/engine code runs.

PR Checklist

Please review the following before submitting your PR:

• PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.

• PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.

• Test cases are provided for new code paths (see test instructions)

• Any new dependencies have been scanned for license and vulnerabilities

• CODEOWNERS updated if ownership changes

• Documentation updated as needed

• Update tava architecture diagram if there is a significant design change in PR.

• The reviewers assigned automatically/manually are appropriate for the PR.

• Please check this after reviewing the above items as appropriate for this PR.

GitHub Bot Help

To see a list of available CI bot commands, please comment /bot help.

Summary by CodeRabbit

• Chores
  • Added license header comments to a utility module.
  • Tweaked an internal quantization list (no change to contents or runtime behavior).

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: c13de3d9-f19d-4b12-83e8-01cd282e15bd

📥 Commits

Reviewing files that changed from the base of the PR and between 36ac0e1 and 32396e7.

📒 Files selected for processing (1)

• tensorrt_llm/bench/utils/__init__.py

✅ Files skipped from review due to trivial changes (1)

• tensorrt_llm/bench/utils/init.py

---

📝 Walkthrough

Walkthrough

Added license header comments and a line-ending edit in tensorrt_llm/bench/utils/__init__.py; VALID_QUANT_ALGOS was edited (removed and re-added an entry) with no net change to its contents or behavior.

Changes

Cohort / File(s)	Summary
Bench utils — minor edit tensorrt_llm/bench/utils/__init__.py	Added SPDX copyright/license header and adjusted a line in the VALID_QUANT_ALGOS literal (removed then re-added QuantAlgo.NVFP4), resulting in no functional or semantic change.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~2 minutes

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately describes the main change: removing INT8 from the quantization choices in trtllm-bench, which is the primary objective of the PR.
Description check	✅ Passed	The PR description comprehensively explains the issue, solution, test coverage, and includes the completed checklist items required by the template.
Linked Issues check	✅ Passed	The PR successfully addresses issue #7091 by removing INT8 from VALID_QUANT_ALGOS, ensuring the CLI rejects INT8 early rather than failing with an AssertionError.
Out of Scope Changes check	✅ Passed	The changes are limited to removing INT8 from VALID_QUANT_ALGOS and adding copyright headers, both directly related to resolving issue #7091.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[Brumbelow]

@coderabbitai review

[coderabbitai]

✅ Actions performed

Review triggered.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

[coderabbitai]

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

tensorrt_llm/bench/utils/__init__.py (1)

1-1: ⚠️ Potential issue | 🟠 Major

Add/update NVIDIA copyright header in this modified source file.

This file appears to be missing the required header (or latest modification year update), which is mandatory for modified TensorRT-LLM source files.

As per coding guidelines, "All TensorRT-LLM source files must contain an NVIDIA copyright header with the year of latest meaningful modification."

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@tensorrt_llm/bench/utils/__init__.py` at line 1, The file
tensorrt_llm/bench/utils/__init__.py is missing the required NVIDIA copyright
header; add the project's canonical header block at the very top of the file
(above the existing import functools), updating the year to the latest
meaningful modification year and preserving the exact template/format used
across the repo, so the header appears before any code or imports in
__init__.py.

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Outside diff comments:
In `@tensorrt_llm/bench/utils/__init__.py`:
- Line 1: The file tensorrt_llm/bench/utils/__init__.py is missing the required
NVIDIA copyright header; add the project's canonical header block at the very
top of the file (above the existing import functools), updating the year to the
latest meaningful modification year and preserving the exact template/format
used across the repo, so the header appears before any code or imports in
__init__.py.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: 65d1134d-d2bb-429d-b1d8-4b2e3b757ad6

📥 Commits

Reviewing files that changed from the base of the PR and between 6e5a339 and 36ac0e1.

📒 Files selected for processing (1)

• tensorrt_llm/bench/utils/__init__.py

[Brumbelow]

@coderabbitai review

[coderabbitai]

✅ Actions performed

Review triggered.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.