From d46afbf8b4bbdcfc8771820d7e881856d66ba7c4 Mon Sep 17 00:00:00 2001 From: Brayden Zhong Date: Wed, 24 Jun 2026 12:34:11 -0700 Subject: [PATCH] Fix nightly CI test for GLM-4.6 + B200 (#28749) Co-authored-by: Brayden Zhong --- python/sglang/srt/server_args.py | 2 +- scripts/ci/utils/slash_command_handler.py | 14 +++++++++----- 2 files changed, 10 insertions(+), 6 deletions(-) diff --git a/python/sglang/srt/server_args.py b/python/sglang/srt/server_args.py index ce2e57b00..caa6967e0 100644 --- a/python/sglang/srt/server_args.py +++ b/python/sglang/srt/server_args.py @@ -4327,7 +4327,7 @@ class ServerArgs: ): self.quantization = quant_method if ( - self.quantization == "modelopt_fp4" + self.quantization in {"modelopt_fp4", None} and self.moe_a2a_backend == "none" and self.moe_runner_backend == "auto" ): diff --git a/scripts/ci/utils/slash_command_handler.py b/scripts/ci/utils/slash_command_handler.py index bbc661493..dc151aa35 100644 --- a/scripts/ci/utils/slash_command_handler.py +++ b/scripts/ci/utils/slash_command_handler.py @@ -725,7 +725,7 @@ _LEGACY_SUITE_TO_RUNNER_CONFIG = { "nightly-perf-vlm-2-gpu": "2-gpu-large", "nightly-4-gpu": "4-gpu-h100", "nightly-4-gpu-b200": "4-gpu-b200", - "nightly-8-gpu-common": "8-gpu-h200", + "nightly-8-gpu-common": ["8-gpu-h200", "8-gpu-b200"], "nightly-8-gpu-h200": "8-gpu-h200", "nightly-kernel-8-gpu-h200": "8-gpu-h200", "nightly-precision-8-gpu-h200": "8-gpu-h200", @@ -830,10 +830,14 @@ def detect_suite(file_path_from_test): legacy_suites = _extract_legacy_suites(content) mappable = [s for s in legacy_suites if s in _LEGACY_SUITE_TO_RUNNER_CONFIG] if mappable: - return [ - _resolve_runner_config(_LEGACY_SUITE_TO_RUNNER_CONFIG[s], full_path, s) - for s in mappable - ] + results = [] + for s in mappable: + rcs = _LEGACY_SUITE_TO_RUNNER_CONFIG[s] + if isinstance(rcs, str): + rcs = [rcs] + for rc in rcs: + results.append(_resolve_runner_config(rc, full_path, s)) + return results if re.search(r"^[^#\n]*register_cpu_ci\s*\(", content, re.MULTILINE): return [