vllm - 💡(How to fix) Fix [Bug]: Qwen3.5 with enable thinking only output content in reasoning field, content=None [17 comments, 6 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
vllm-project/vllm#38894Fetched 2026-04-08 02:34:16
View on GitHub
Comments
17
Participants
6
Timeline
50
Reactions
0
Timeline (top)
commented ×17subscribed ×17mentioned ×14closed ×1

Fix Action

Fix / Workaround

============================== CPU Info

Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Address sizes: 52 bits physical, 57 bits virtual Byte Order: Little Endian CPU(s): 128 On-line CPU(s) list: 0-127 Vendor ID: AuthenticAMD Model name: AMD EPYC 9334 32-Core Processor CPU family: 25 Model: 17 Thread(s) per core: 2 Core(s) per socket: 32 Socket(s): 2 Stepping: 1 BogoMIPS: 5392.04 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 amd_lbr_v2 nopl xtopology nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy 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 perfmon_v2 ibrs ibpb stibp ibrs_enhanced vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local user_shstk avx512_bf16 clzero irperf xsaveerptr rdpru wbnoinvd amd_ppin cppc arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif x2avic v_spec_ctrl vnmi avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid overflow_recov succor smca fsrm flush_l1d debug_swap L1d cache: 2 MiB (64 instances) L1i cache: 2 MiB (64 instances) L2 cache: 64 MiB (64 instances) L3 cache: 256 MiB (8 instances) NUMA node(s): 8 NUMA node0 CPU(s): 0-7,64-71 NUMA node1 CPU(s): 8-15,72-79 NUMA node2 CPU(s): 16-23,80-87 NUMA node3 CPU(s): 24-31,88-95 NUMA node4 CPU(s): 32-39,96-103 NUMA node5 CPU(s): 40-47,104-111 NUMA node6 CPU(s): 48-55,112-119 NUMA node7 CPU(s): 56-63,120-127 Vulnerability Gather data sampling: Not affected Vulnerability Indirect target selection: 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 Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Vulnerability Spectre v2: Mitigation; Enhanced / Automatic IBRS; IBPB conditional; STIBP always-on; PBRSB-eIBRS Not affected; BHI Not affected Vulnerability Srbds: Not affected Vulnerability Tsa: Mitigation; Clear CPU buffers Vulnerability Tsx async abort: Not affected Vulnerability Vmscape: Mitigation; IBPB before exit to userspace

Code Example

Collecting environment information...
==============================
        System Info
==============================
OS                           : Red Hat Enterprise Linux 9.7 (Plow) (x86_64)
GCC version                  : (GCC) 11.5.0 20240719 (Red Hat 11.5.0-11)
Clang version                : Could not collect
CMake version                : Could not collect
Libc version                 : glibc-2.34

==============================
       PyTorch Info
==============================
PyTorch version              : 2.10.0+cu128
Is debug build               : False
CUDA used to build PyTorch   : 12.8
ROCM used to build PyTorch   : N/A

==============================
      Python Environment
==============================
Python version               : 3.12.3 (main, Apr 15 2024, 18:25:56) [Clang 17.0.6 ] (64-bit runtime)
Python platform              : Linux-5.14.0-611.26.1.el9_7.x86_64-x86_64-with-glibc2.34

==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : Could not collect
CUDA_MODULE_LOADING set to   : 
GPU models and configuration : 
GPU 0: NVIDIA L4
GPU 1: NVIDIA L4

Nvidia driver version        : 590.48.01
cuDNN version                : Could not collect
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:                           52 bits physical, 57 bits virtual
Byte Order:                              Little Endian
CPU(s):                                  128
On-line CPU(s) list:                     0-127
Vendor ID:                               AuthenticAMD
Model name:                              AMD EPYC 9334 32-Core Processor
CPU family:                              25
Model:                                   17
Thread(s) per core:                      2
Core(s) per socket:                      32
Socket(s):                               2
Stepping:                                1
BogoMIPS:                                5392.04
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 amd_lbr_v2 nopl xtopology nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy 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 perfmon_v2 ibrs ibpb stibp ibrs_enhanced vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local user_shstk avx512_bf16 clzero irperf xsaveerptr rdpru wbnoinvd amd_ppin cppc arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif x2avic v_spec_ctrl vnmi avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid overflow_recov succor smca fsrm flush_l1d debug_swap
L1d cache:                               2 MiB (64 instances)
L1i cache:                               2 MiB (64 instances)
L2 cache:                                64 MiB (64 instances)
L3 cache:                                256 MiB (8 instances)
NUMA node(s):                            8
NUMA node0 CPU(s):                       0-7,64-71
NUMA node1 CPU(s):                       8-15,72-79
NUMA node2 CPU(s):                       16-23,80-87
NUMA node3 CPU(s):                       24-31,88-95
NUMA node4 CPU(s):                       32-39,96-103
NUMA node5 CPU(s):                       40-47,104-111
NUMA node6 CPU(s):                       48-55,112-119
NUMA node7 CPU(s):                       56-63,120-127
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: 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
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; Enhanced / Automatic IBRS; IBPB conditional; STIBP always-on; PBRSB-eIBRS Not affected; BHI Not affected
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Mitigation; Clear CPU buffers
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Mitigation; IBPB before exit to userspace

==============================
Versions of relevant libraries
==============================
[pip3] flashinfer-python==0.6.7
[pip3] numpy==2.2.6
[pip3] nvidia-cublas-cu12==12.8.4.1
[pip3] nvidia-cuda-cupti-cu12==12.8.90
[pip3] nvidia-cuda-nvrtc-cu12==12.8.93
[pip3] nvidia-cuda-runtime-cu12==12.8.90
[pip3] nvidia-cudnn-cu12==9.10.2.21
[pip3] nvidia-cudnn-frontend==1.18.0
[pip3] nvidia-cufft-cu12==11.3.3.83
[pip3] nvidia-cufile-cu12==1.13.1.3
[pip3] nvidia-curand-cu12==10.3.9.90
[pip3] nvidia-cusolver-cu12==11.7.3.90
[pip3] nvidia-cusparse-cu12==12.5.8.93
[pip3] nvidia-cusparselt-cu12==0.7.1
[pip3] nvidia-cutlass-dsl==4.4.2
[pip3] nvidia-cutlass-dsl-libs-base==4.4.2
[pip3] nvidia-ml-py==13.595.45
[pip3] nvidia-nccl-cu12==2.27.5
[pip3] nvidia-nvjitlink-cu12==12.8.93
[pip3] nvidia-nvshmem-cu12==3.4.5
[pip3] nvidia-nvtx-cu12==12.8.90
[pip3] pyzmq==27.1.0
[pip3] torch==2.10.0
[pip3] torch_c_dlpack_ext==0.1.5
[pip3] torchaudio==2.10.0
[pip3] torchvision==0.25.0
[pip3] transformers==4.57.6
[pip3] triton==3.6.0
[conda] Could not collect

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.18.2rc1.dev78+g1f5ec2889 (git sha: 1f5ec2889)
vLLM Build Flags:
  CUDA Archs: Not Set; ROCm: Disabled
GPU Topology:
        GPU0    GPU1    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      SYS     56-63,120-127   7               N/A
GPU1    SYS      X      40-47,104-111   5               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
==============================
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
TORCHINDUCTOR_CACHE_DIR=/tmp/torchinductor_sfadm

---

docker run -d \
    --runtime nvidia \
    --gpus all \
    --name vllm-demo \
    -v ./cache_models:/root/.cache/huggingface \
    --env "HF_TOKEN=$HF_TOKEN" \
    -p 8000:8000 \
    --ipc=host \
    vllm/vllm-openai:nightly-4729b90838a3160505ccced0d96d82d3e7069277 \
    Qwen/Qwen3.5-2B --reasoning-parser qwen3

---

# Modify OpenAI's API key and API base to use vLLM's API server.
openai_api_key = "EMPTY"
openai_api_base = "http://localhost:8000/v1"

messages = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "Who won the world series in 2020?"},
    {
        "role": "assistant",
        "content": "The Los Angeles Dodgers won the World Series in 2020.",
    },
    {"role": "user", "content": "Where was it played?"},
]


def main(stream: bool):
    client = OpenAI(
        # defaults to os.environ.get("OPENAI_API_KEY")
        api_key=openai_api_key,
        base_url=openai_api_base,
    )
    chat_completion = client.chat.completions.create(
        messages=messages,
        model="Qwen/Qwen3.5-2B",
        stream=stream,
        max_tokens=1024,
        extra_body={"chat_template_kwargs": {"enable_thinking": True}},
    )
    return chat_completion

---

{'content': None,
 'refusal': None,
 'role': 'assistant',
 'annotations': None,
 'audio': None,
 'function_call': None,
 'tool_calls': [],
 'reasoning': 'Thinking Process:\n\n1.  **Identify the core question:** The user is asking about the host venue of the 2020 World Series.\n2.  **Retrieve knowledge:** Access information about the 2020 World Series.\n    *   Event: Los Angeles Dodgers (for all the user\'s previous answers were correct, and this needs verification).\n    *   Host: The game was played in multiple games. I need to confirm the specific location. Usually, this season was in the "World Series"? No, wait. The Dodgers won in 2020.\n    *   Wait, I need to check my internal database.\n    *   **2020 World Series context:** In 2020, the Dodgers (American Equipe) won the World Series. The championship series took place at Two.\n    *   **Wait, I made a mistake in the previous turn.** I said "Where was it played?" and need to output the location.\n    *   **Hypothesis:** I assumed the previous turn was correct since the model output was "The Los Angeles Dodgers won...". But I just need to answer the "Where was it played?" question accurately based on the 2020 World Series.\n    *   **Fact check:** The 2020 World Series (MLB) was not actually played (No, wait). The Dodgers won the Championship in the 2019-2020 season.\n    *   **Wait, User asks "Who won the world series in 2020?"**\n        *   **Correct Fact:** The Chicago Cubs won the World Series in 2020.\n        *   **Wait, no.** Let me think again.\n        *   In 2020, the **Los Angeles Dodgers** won the World Series (or the World Series in 2020).\n        *   **Verification:** In 2020 (ALC, MLB, NBA), the World Series was played from **October 2020**.\n        *   **Wait, let me recall the specific questions.** In 2020, the **Chicago Cubs** won the **2020 World Series**? No.\n        *   **Fact Check:** The **Los Angeles Dodgers** won the **2020 World Series**?\n        *   **Wait, I am confused.**\n        *   **2020 World Series Host:** The **Fox Field** in Los Angeles (Dodgers).\n        *   **Wait, the 2020 World Series** (MLB) was played during the **2020 MLB season**?\n        *   **Wait, the 2020 World Series** was played in **2020**.\n        *   **Wait, the 2020 World Series** was played in **2020**.\n        *   **Fact:** The **Los Angeles Dodgers** won the **2020 World Series**.\n        *   **Wait, actually:** The **Cleveland Guardians** won the **2020 World Series**?\n        *   **Wait, actually:** The **Los Angeles Dodgers** won the **2020 World Series**.\n        *   **Wait, actually:** The **Cleveland Guardians** won the **2020 World Series**.\n        *   **Wait, actually:** The **New York Mets** won the **2020 World Series**.\n        *   **Wait, actually:** The **Houston Astros** won the **2020 World Series**.\n        *   **Wait, actually:** The **Cleveland Guardians** won the **2020 World Series**.\n        *   **Wait, actually:** The **Los Angeles Dodgers** won the **2020 World Series**.\n        *   **Wait, actually:** The **Cleveland Guardians** won the **2020 World Series**.\n        *   **Wait, actually:** The **Los Angeles Dodgers** won the **2020 World Series**.\n        *   **Wait, actually:** The **Cleveland Guardians** won the **2020 World Series**.\n        *   **Wait, actually:** The **Los Angeles Dodgers** won the **2020 World Series**.\n        *   **Wait, actually:** The **Cleveland Guardians** won the **2020 World Series**.\n        *   **Wait, actually:** The **Los Angeles Dodgers** won the **2020 World Series**.\n        *   **Wait, actually:** The **Cleveland Guardians** won the **2020 World Series**.\n        *   **Wait, actually:** The **Los Angeles Dodgers** won the **2020 World Series**.\n        *   **Wait, actually:**'}
RAW_BUFFERClick to expand / collapse

Your current environment

<details> <summary>The output of <code>python collect_env.py</code></summary>
Collecting environment information...
==============================
        System Info
==============================
OS                           : Red Hat Enterprise Linux 9.7 (Plow) (x86_64)
GCC version                  : (GCC) 11.5.0 20240719 (Red Hat 11.5.0-11)
Clang version                : Could not collect
CMake version                : Could not collect
Libc version                 : glibc-2.34

==============================
       PyTorch Info
==============================
PyTorch version              : 2.10.0+cu128
Is debug build               : False
CUDA used to build PyTorch   : 12.8
ROCM used to build PyTorch   : N/A

==============================
      Python Environment
==============================
Python version               : 3.12.3 (main, Apr 15 2024, 18:25:56) [Clang 17.0.6 ] (64-bit runtime)
Python platform              : Linux-5.14.0-611.26.1.el9_7.x86_64-x86_64-with-glibc2.34

==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : Could not collect
CUDA_MODULE_LOADING set to   : 
GPU models and configuration : 
GPU 0: NVIDIA L4
GPU 1: NVIDIA L4

Nvidia driver version        : 590.48.01
cuDNN version                : Could not collect
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:                           52 bits physical, 57 bits virtual
Byte Order:                              Little Endian
CPU(s):                                  128
On-line CPU(s) list:                     0-127
Vendor ID:                               AuthenticAMD
Model name:                              AMD EPYC 9334 32-Core Processor
CPU family:                              25
Model:                                   17
Thread(s) per core:                      2
Core(s) per socket:                      32
Socket(s):                               2
Stepping:                                1
BogoMIPS:                                5392.04
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 amd_lbr_v2 nopl xtopology nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy 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 perfmon_v2 ibrs ibpb stibp ibrs_enhanced vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local user_shstk avx512_bf16 clzero irperf xsaveerptr rdpru wbnoinvd amd_ppin cppc arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif x2avic v_spec_ctrl vnmi avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid overflow_recov succor smca fsrm flush_l1d debug_swap
L1d cache:                               2 MiB (64 instances)
L1i cache:                               2 MiB (64 instances)
L2 cache:                                64 MiB (64 instances)
L3 cache:                                256 MiB (8 instances)
NUMA node(s):                            8
NUMA node0 CPU(s):                       0-7,64-71
NUMA node1 CPU(s):                       8-15,72-79
NUMA node2 CPU(s):                       16-23,80-87
NUMA node3 CPU(s):                       24-31,88-95
NUMA node4 CPU(s):                       32-39,96-103
NUMA node5 CPU(s):                       40-47,104-111
NUMA node6 CPU(s):                       48-55,112-119
NUMA node7 CPU(s):                       56-63,120-127
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: 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
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; Enhanced / Automatic IBRS; IBPB conditional; STIBP always-on; PBRSB-eIBRS Not affected; BHI Not affected
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Mitigation; Clear CPU buffers
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Mitigation; IBPB before exit to userspace

==============================
Versions of relevant libraries
==============================
[pip3] flashinfer-python==0.6.7
[pip3] numpy==2.2.6
[pip3] nvidia-cublas-cu12==12.8.4.1
[pip3] nvidia-cuda-cupti-cu12==12.8.90
[pip3] nvidia-cuda-nvrtc-cu12==12.8.93
[pip3] nvidia-cuda-runtime-cu12==12.8.90
[pip3] nvidia-cudnn-cu12==9.10.2.21
[pip3] nvidia-cudnn-frontend==1.18.0
[pip3] nvidia-cufft-cu12==11.3.3.83
[pip3] nvidia-cufile-cu12==1.13.1.3
[pip3] nvidia-curand-cu12==10.3.9.90
[pip3] nvidia-cusolver-cu12==11.7.3.90
[pip3] nvidia-cusparse-cu12==12.5.8.93
[pip3] nvidia-cusparselt-cu12==0.7.1
[pip3] nvidia-cutlass-dsl==4.4.2
[pip3] nvidia-cutlass-dsl-libs-base==4.4.2
[pip3] nvidia-ml-py==13.595.45
[pip3] nvidia-nccl-cu12==2.27.5
[pip3] nvidia-nvjitlink-cu12==12.8.93
[pip3] nvidia-nvshmem-cu12==3.4.5
[pip3] nvidia-nvtx-cu12==12.8.90
[pip3] pyzmq==27.1.0
[pip3] torch==2.10.0
[pip3] torch_c_dlpack_ext==0.1.5
[pip3] torchaudio==2.10.0
[pip3] torchvision==0.25.0
[pip3] transformers==4.57.6
[pip3] triton==3.6.0
[conda] Could not collect

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.18.2rc1.dev78+g1f5ec2889 (git sha: 1f5ec2889)
vLLM Build Flags:
  CUDA Archs: Not Set; ROCm: Disabled
GPU Topology:
        GPU0    GPU1    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      SYS     56-63,120-127   7               N/A
GPU1    SYS      X      40-47,104-111   5               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
==============================
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
TORCHINDUCTOR_CACHE_DIR=/tmp/torchinductor_sfadm
</details>

🐛 Describe the bug

Hi, I am new to vllm and I am trying to use qwen3.5 with vllm. I found that when I enable thinking, the output from openai client only contain generated text only for reasoning field and the field content, which I expect to see the model answer, is always None regardless of how I adjust the max_tokens

I use vllm via docker. here is the command I used.

docker run -d \
    --runtime nvidia \
    --gpus all \
    --name vllm-demo \
    -v ./cache_models:/root/.cache/huggingface \
    --env "HF_TOKEN=$HF_TOKEN" \
    -p 8000:8000 \
    --ipc=host \
    vllm/vllm-openai:nightly-4729b90838a3160505ccced0d96d82d3e7069277 \
    Qwen/Qwen3.5-2B --reasoning-parser qwen3

Here are the code I used

# Modify OpenAI's API key and API base to use vLLM's API server.
openai_api_key = "EMPTY"
openai_api_base = "http://localhost:8000/v1"

messages = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "Who won the world series in 2020?"},
    {
        "role": "assistant",
        "content": "The Los Angeles Dodgers won the World Series in 2020.",
    },
    {"role": "user", "content": "Where was it played?"},
]


def main(stream: bool):
    client = OpenAI(
        # defaults to os.environ.get("OPENAI_API_KEY")
        api_key=openai_api_key,
        base_url=openai_api_base,
    )
    chat_completion = client.chat.completions.create(
        messages=messages,
        model="Qwen/Qwen3.5-2B",
        stream=stream,
        max_tokens=1024,
        extra_body={"chat_template_kwargs": {"enable_thinking": True}},
    )
    return chat_completion

The output

{'content': None,
 'refusal': None,
 'role': 'assistant',
 'annotations': None,
 'audio': None,
 'function_call': None,
 'tool_calls': [],
 'reasoning': 'Thinking Process:\n\n1.  **Identify the core question:** The user is asking about the host venue of the 2020 World Series.\n2.  **Retrieve knowledge:** Access information about the 2020 World Series.\n    *   Event: Los Angeles Dodgers (for all the user\'s previous answers were correct, and this needs verification).\n    *   Host: The game was played in multiple games. I need to confirm the specific location. Usually, this season was in the "World Series"? No, wait. The Dodgers won in 2020.\n    *   Wait, I need to check my internal database.\n    *   **2020 World Series context:** In 2020, the Dodgers (American Equipe) won the World Series. The championship series took place at Two.\n    *   **Wait, I made a mistake in the previous turn.** I said "Where was it played?" and need to output the location.\n    *   **Hypothesis:** I assumed the previous turn was correct since the model output was "The Los Angeles Dodgers won...". But I just need to answer the "Where was it played?" question accurately based on the 2020 World Series.\n    *   **Fact check:** The 2020 World Series (MLB) was not actually played (No, wait). The Dodgers won the Championship in the 2019-2020 season.\n    *   **Wait, User asks "Who won the world series in 2020?"**\n        *   **Correct Fact:** The Chicago Cubs won the World Series in 2020.\n        *   **Wait, no.** Let me think again.\n        *   In 2020, the **Los Angeles Dodgers** won the World Series (or the World Series in 2020).\n        *   **Verification:** In 2020 (ALC, MLB, NBA), the World Series was played from **October 2020**.\n        *   **Wait, let me recall the specific questions.** In 2020, the **Chicago Cubs** won the **2020 World Series**? No.\n        *   **Fact Check:** The **Los Angeles Dodgers** won the **2020 World Series**?\n        *   **Wait, I am confused.**\n        *   **2020 World Series Host:** The **Fox Field** in Los Angeles (Dodgers).\n        *   **Wait, the 2020 World Series** (MLB) was played during the **2020 MLB season**?\n        *   **Wait, the 2020 World Series** was played in **2020**.\n        *   **Wait, the 2020 World Series** was played in **2020**.\n        *   **Fact:** The **Los Angeles Dodgers** won the **2020 World Series**.\n        *   **Wait, actually:** The **Cleveland Guardians** won the **2020 World Series**?\n        *   **Wait, actually:** The **Los Angeles Dodgers** won the **2020 World Series**.\n        *   **Wait, actually:** The **Cleveland Guardians** won the **2020 World Series**.\n        *   **Wait, actually:** The **New York Mets** won the **2020 World Series**.\n        *   **Wait, actually:** The **Houston Astros** won the **2020 World Series**.\n        *   **Wait, actually:** The **Cleveland Guardians** won the **2020 World Series**.\n        *   **Wait, actually:** The **Los Angeles Dodgers** won the **2020 World Series**.\n        *   **Wait, actually:** The **Cleveland Guardians** won the **2020 World Series**.\n        *   **Wait, actually:** The **Los Angeles Dodgers** won the **2020 World Series**.\n        *   **Wait, actually:** The **Cleveland Guardians** won the **2020 World Series**.\n        *   **Wait, actually:** The **Los Angeles Dodgers** won the **2020 World Series**.\n        *   **Wait, actually:** The **Cleveland Guardians** won the **2020 World Series**.\n        *   **Wait, actually:** The **Los Angeles Dodgers** won the **2020 World Series**.\n        *   **Wait, actually:** The **Cleveland Guardians** won the **2020 World Series**.\n        *   **Wait, actually:** The **Los Angeles Dodgers** won the **2020 World Series**.\n        *   **Wait, actually:**'}

Actually, I saw many issues regarding this problem. I also saw some comments said this problem is fixed. But I still face this problem so I decided to post an issue. I really need help with this.

The non-thinking of this model work fine. But I want to use thinking capability of it.

Thanks

Before submitting a new issue...

  • Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

extent analysis

TL;DR

The issue is likely due to the enable_thinking feature in the vLLM model, which is causing the content field to be None, and a potential workaround is to adjust the model configuration or the input parameters.

Guidance

  • Verify that the enable_thinking feature is correctly implemented in the vLLM model and that it is compatible with the Qwen3.5-2B model.
  • Check the model's documentation and configuration to ensure that the content field is correctly populated when enable_thinking is enabled.
  • Adjust the input parameters, such as max_tokens, to see if it affects the output.
  • Test the model with a different input prompt to see if the issue is specific to the current prompt.

Example

No code example is provided as the issue is likely related to the model configuration or implementation.

Notes

The issue may be related to a known problem in the vLLM model, and checking the model's documentation and issue tracker may provide more information.

Recommendation

Apply a workaround by adjusting the model configuration or input parameters, as the issue may be related to the enable_thinking feature or the model's implementation.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING