Checklist

• 1. I have searched related issues but cannot get the expected help.
• 2. The bug has not been fixed in the latest version.
• 3. Please note that if the bug-related issue you submitted lacks corresponding environment info and a minimal reproducible demo, it will be challenging for us to reproduce and resolve the issue, reducing the likelihood of receiving feedback.
• 4. If the issue you raised is not a bug but a question, please raise a discussion at https://github.com/sgl-project/sglang/discussions/new/choose Otherwise, it will be closed.
• 5. Please use English, otherwise it will be closed.

Describe the bug

Both Qwen3-235B-A22B-Thinking-FP8 and Qwen3-235B-A22B-Thinking models are not deterministic even with --enable-deterministic-inference.

Reproduction

# H200 x 4 Qwen3-235B-A22B-Thinking-2507-FP8
python3 -m sglang.launch_server --model-path /shared/public/elr-models/Qwen/Qwen3-235B-A22B-Thinking-2507-FP8/f07f63f2bbd7540917118ebdf3812696ef303b03/ --attention-backend fa3 --enable-deterministic-inference --tp 4

python3 -m sglang.test.test_deterministic --test-mode prefix --n-trials 50
Prompt 0 with prefix length 1: total samples: 313, Unique samples: 2
Prompt 1 with prefix length 511: total samples: 313, Unique samples: 5
Prompt 2 with prefix length 2048: total samples: 303, Unique samples: 6
Prompt 3 with prefix length 4097: total samples: 346, Unique samples: 8

# H200 x 8 Qwen3-235B-A22B-Thinking-2507
python3 -m sglang.launch_server --model-path /shared/public/elr-models/Qwen/Qwen3-235B-A22B-Thinking-2507/ddee1c5fb2c5406e8315f8a6d47214c74349e15b/ --attention-backend fa3 --enable-deterministic-inference --tp 8

python3 -m sglang.test.test_deterministic --test-mode prefix --n-trials 50
Prompt 0 with prefix length 1: total samples: 292, Unique samples: 1
Prompt 1 with prefix length 511: total samples: 321, Unique samples: 1
Prompt 2 with prefix length 2048: total samples: 330, Unique samples: 2
Prompt 3 with prefix length 4097: total samples: 332, Unique samples: 2

Environment

Python: 3.10.12 (main, May 27 2025, 17:12:29) [GCC 11.4.0]
CUDA available: True
GPU 0,1,2,3,4,5,6,7: NVIDIA H200
GPU 0,1,2,3,4,5,6,7 Compute Capability: 9.0
CUDA_HOME: /usr/local/cuda
NVCC: Cuda compilation tools, release 12.8, V12.8.93
CUDA Driver Version: 550.163.01
PyTorch: 2.7.1+cu126
sglang: 0.5.4
sgl_kernel: 0.2.7
flashinfer_python: 0.2.9rc2
triton: 3.3.1
transformers: 4.54.0
torchao: 0.9.0+cu126
numpy: 1.24.3
aiohttp: 3.12.15
fastapi: 0.116.1
hf_transfer: 0.1.9
huggingface_hub: 0.34.3
interegular: 0.3.3
modelscope: 1.28.1
orjson: 3.11.1
outlines: 0.1.11
packaging: 25.0
psutil: 7.0.0
pydantic: 2.11.7
python-multipart: 0.0.20
pyzmq: 27.0.0
uvicorn: 0.35.0
uvloop: 0.21.0
vllm: Module Not Found
xgrammar: 0.1.21
openai: 1.97.1
tiktoken: 0.9.0
anthropic: 0.60.0
litellm: 1.74.9.post1
decord2: Module Not Found
NVIDIA Topology:
GPU0 GPU1 GPU2 GPU3 GPU4 GPU5 GPU6 GPU7 NIC0 NIC1 NIC2 NIC3 NIC4 NIC5 CPU Affinity NUMA Affinity GPU NUMA ID
GPU0 X NV18 NV18 NV18 NV18 NV18 NV18 NV18 PIX NODE NODE NODE SYS SYS 0-63,128-191 0 N/A
GPU1 NV18 X NV18 NV18 NV18 NV18 NV18 NV18 NODE PHB PHB NODE SYS SYS 0-63,128-191 0 N/A
GPU2 NV18 NV18 X NV18 NV18 NV18 NV18 NV18 NODE NODE NODE PIX SYS SYS 0-63,128-191 0 N/A
GPU3 NV18 NV18 NV18 X NV18 NV18 NV18 NV18 NODE NODE NODE NODE SYS SYS 0-63,128-191 0 N/A
GPU4 NV18 NV18 NV18 NV18 X NV18 NV18 NV18 SYS SYS SYS SYS PIX NODE 64-127,192-255 1 N/A
GPU5 NV18 NV18 NV18 NV18 NV18 X NV18 NV18 SYS SYS SYS SYS NODE NODE 64-127,192-255 1 N/A
GPU6 NV18 NV18 NV18 NV18 NV18 NV18 X NV18 SYS SYS SYS SYS NODE PIX 64-127,192-255 1 N/A
GPU7 NV18 NV18 NV18 NV18 NV18 NV18 NV18 X SYS SYS SYS SYS NODE NODE 64-127,192-255 1 N/A
NIC0 PIX NODE NODE NODE SYS SYS SYS SYS X NODE NODE NODE SYS SYS
NIC1 NODE PHB NODE NODE SYS SYS SYS SYS NODE X PIX NODE SYS SYS
NIC2 NODE PHB NODE NODE SYS SYS SYS SYS NODE PIX X NODE SYS SYS
NIC3 NODE NODE PIX NODE SYS SYS SYS SYS NODE NODE NODE X SYS SYS
NIC4 SYS SYS SYS SYS PIX NODE NODE NODE SYS SYS SYS SYS X NODE
NIC5 SYS SYS SYS SYS NODE NODE PIX NODE SYS SYS SYS SYS NODE X

Legend:

X = Self
SYS = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)
NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA node
PHB = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)
PXB = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)
PIX = Connection traversing at most a single PCIe bridge
NV# = Connection traversing a bonded set of # NVLinks

NIC Legend:

NIC0: mlx5_0
NIC1: mlx5_1
NIC2: mlx5_2
NIC3: mlx5_3
NIC4: mlx5_4
NIC5: mlx5_5

ulimit soft: 10000000