vllm
vllm copied to clipboard
Published
20 hours ago •
vllm-project
vllm-project
[Bug]: OpenAI Classification Client returning logits instead of softmax values
environment info
### Your current environment
==============================
System Info
==============================
OS : Ubuntu 22.04.4 LTS (x86_64)
GCC version : (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version : Could not collect
CMake version : version 3.22.1
Libc version : glibc-2.35
==============================
PyTorch Info
==============================
PyTorch version : 2.7.0+cu126
Is debug build : False
CUDA used to build PyTorch : 12.6
ROCM used to build PyTorch : N/A
==============================
Python Environment
==============================
Python version : 3.10.12 (main, Feb 4 2025, 14:57:36) [GCC 11.4.0] (64-bit runtime)
Python platform : Linux-5.15.0-139-generic-x86_64-with-glibc2.35
==============================
CUDA / GPU Info
==============================
Is CUDA available : True
CUDA runtime version : 11.5.119
CUDA_MODULE_LOADING set to : LAZY
GPU models and configuration :
GPU 0: NVIDIA GeForce RTX 3090
GPU 1: NVIDIA GeForce RTX 3090
Nvidia driver version : 535.230.02
cuDNN version : Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_adv.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_cnn.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_engines_precompiled.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_engines_runtime_compiled.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_graph.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_heuristic.so.9.2.1
/usr/lib/x86_64-linux-gnu/libcudnn_ops.so.9.2.1
HIP runtime version : N/A
MIOpen runtime version : N/A
Is XNNPACK available : True
==============================
CPU Info
==============================
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 48 bits physical, 48 bits virtual
Byte Order: Little Endian
CPU(s): 24
On-line CPU(s) list: 0-23
Vendor ID: AuthenticAMD
Model name: AMD Ryzen 9 5900X 12-Core Processor
CPU family: 25
Model: 33
Thread(s) per core: 2
Core(s) per socket: 12
Socket(s): 1
Stepping: 0
Frequency boost: enabled
CPU max MHz: 3700.0000
CPU min MHz: 2200.0000
BogoMIPS: 7385.88
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate ssbd mba ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr rdpru wbnoinvd arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif v_spec_ctrl umip pku ospke vaes vpclmulqdq rdpid overflow_recov succor smca fsrm
Virtualization: AMD-V
L1d cache: 384 KiB (12 instances)
L1i cache: 384 KiB (12 instances)
L2 cache: 6 MiB (12 instances)
L3 cache: 64 MiB (2 instances)
NUMA node(s): 1
NUMA node0 CPU(s): 0-23
Vulnerability Gather data sampling: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Mitigation; safe RET, no microcode
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; Retpolines; IBPB conditional; IBRS_FW; STIBP always-on; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Not affected
==============================
Versions of relevant libraries
==============================
[pip3] numpy==2.2.6
[pip3] nvidia-cublas-cu12==12.6.4.1
[pip3] nvidia-cuda-cupti-cu12==12.6.80
[pip3] nvidia-cuda-nvrtc-cu12==12.6.77
[pip3] nvidia-cuda-runtime-cu12==12.6.77
[pip3] nvidia-cudnn-cu12==9.5.1.17
[pip3] nvidia-cufft-cu12==11.3.0.4
[pip3] nvidia-cufile-cu12==1.11.1.6
[pip3] nvidia-curand-cu12==10.3.7.77
[pip3] nvidia-cusolver-cu12==11.7.1.2
[pip3] nvidia-cusparse-cu12==12.5.4.2
[pip3] nvidia-cusparselt-cu12==0.6.3
[pip3] nvidia-nccl-cu12==2.26.2
[pip3] nvidia-nvjitlink-cu12==12.6.85
[pip3] nvidia-nvtx-cu12==12.6.77
[pip3] pyzmq==26.4.0
[pip3] torch==2.7.0
[pip3] torchaudio==2.7.0
[pip3] torchvision==0.22.0
[pip3] transformers==4.52.3
[pip3] triton==3.3.0
[conda] Could not collect
==============================
vLLM Info
==============================
ROCM Version : Could not collect
Neuron SDK Version : N/A
vLLM Version : 0.9.0
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0 GPU1 CPU Affinity NUMA Affinity GPU NUMA ID
GPU0 X PHB 0-23 0 N/A
GPU1 PHB X 0-23 0 N/A
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
==============================
Environment Variables
==============================
NCCL_CUMEM_ENABLE=0
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
CUDA_MODULE_LOADING=LAZY
🐛 Describe the bug
After installing 0.9.0 pre-release from the following wheel https://wheels.vllm.ai/c533c6fa8403f8bb3a75f130e063dbaafc1d69dc/vllm-0.9.0-cp38-abi3-manylinux1_x86_64.whl, I see that for papluca/xlm-roberta-base-language-detection model raw logits are being returned instead of softmax output.
Steps to reproduce
pip install https://wheels.vllm.ai/c533c6fa8403f8bb3a75f130e063dbaafc1d69dc/vllm-0.9.0-cp38-abi3-manylinux1_x86_64.whl
vllm serve 'papluca/xlm-roberta-base-language-detection' --port 8098 --host 0.0.0.0 --task classify
Wrong output
When calling the model, I see raw logits being returned:
curl -v "http://127.0.0.1:8098/classify" \
-H "Content-Type: application/json" \
-d '{
"model": "papluca/xlm-roberta-base-language-detection",
"input": "Hello"
}'
Returns:
{
"id": "classify-5b23c5f8efa24ccfb1ca60f6b4283006",
"object": "list",
"created": 1748283054,
"model": "papluca/xlm-roberta-base-language-detection",
"data": [
{
"index": 0,
"label": "en",
"probs": [
-1.0625,
0.290771484375,
-1.125,
-0.9794921875,
-0.339111328125,
-0.1148681640625,
-0.75439453125,
0.360107421875,
-0.479736328125,
1.2421875,
-0.338623046875,
0.671875,
-0.04315185546875,
3.888671875,
-0.52099609375,
-0.26416015625,
0.2705078125,
-0.152099609375,
0.71484375,
-0.5517578125
],
"num_classes": 20
}
],
"usage": {
"prompt_tokens": 3,
"total_tokens": 3,
"completion_tokens": 0,
"prompt_tokens_details": null
}
}
When trying other the model from the docs jason9693/Qwen2.5-1.5B-apeach, which has the same task of SequenceClassification as papluca/xlm-roberta-base-language-detection, seems to be returning the softmax output properly.
Seems like the same happens for offline inference, and it not just an online inference problem.
