Issue Station
Sign In
Tools hereToolbox

vllm - ✅(Solved) Fix [Bug]: /v1/responses: Protocol drift and malformed tool aggregation breaking official OpenAI SDK compatibility [1 pull requests, 3 comments, 3 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#39426•Fetched 2026-04-10 03:40:44
View on GitHub
Comments
3
Participants
3
Timeline
8
Reactions
0
Author
scott8scott8
Participants
chaunceyjiangchaunceyjiangianliuyianliuyscott8scott8
Assignees
chaunceyjiangchaunceyjiang
Timeline (top)
commented ×3referenced ×2assigned ×1cross-referenced ×1

Root Cause

Here’s what appears to be some legitimate vLLM errors. I have a custom AI client using the Python openai library. Gemini 3 Flash (Preview) generated this after working with me on my client implementation. The context for finding this is that we are trying to make my client work with the official OpenAI library and vLLM’s /v1/responses endpoint. We currently have to use a lower-level manual SSE parser to handle the stream because vLLM’s output is incompatible with the SDK’s internal assertions.

Fix Action

Fix / Workaround

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

Architecture: aarch64 CPU op-mode(s): 64-bit Byte Order: Little Endian CPU(s): 20 On-line CPU(s) list: 0-19 Vendor ID: ARM Model name: Cortex-X925 Model: 1 Thread(s) per core: 1 Core(s) per socket: 10 Socket(s): 1 Stepping: r0p1 Frequency boost: disabled CPU(s) scaling MHz: 100% CPU max MHz: 3900.0000 CPU min MHz: 1378.0000 BogoMIPS: 2000.00 Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh bti ecv afp wfxt Model name: Cortex-A725 Model: 1 Thread(s) per core: 1 Core(s) per socket: 10 Socket(s): 1 Stepping: r0p1 CPU(s) scaling MHz: 100% CPU max MHz: 2808.0000 CPU min MHz: 338.0000 BogoMIPS: 2000.00 Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh bti ecv afp wfxt L1d cache: 1.3 MiB (20 instances) L1i cache: 1.3 MiB (20 instances) L2 cache: 25 MiB (20 instances) L3 cache: 24 MiB (2 instances) NUMA node(s): 1 NUMA node0 CPU(s): 0-19 Vulnerability Gather data sampling: Not affected Vulnerability Ghostwrite: 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 Old microcode: Not affected Vulnerability Reg file data sampling: Not affected Vulnerability Retbleed: Not affected Vulnerability Spec rstack overflow: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl Vulnerability Spectre v1: Mitigation; __user pointer sanitization Vulnerability Spectre v2: Mitigation; CSV2, BHB Vulnerability Srbds: Not affected Vulnerability Tsa: Not affected Vulnerability Tsx async abort: Not affected Vulnerability Vmscape: Not affected

Impact: Invented event names like reasoning_part cause drift from the official SDK’s event-dispatch registry and trigger internal assertions during stream parsing.

PR fix notes

PR #39447: fix(responses): fix malformed JSON in multi-tool-call streaming

Description (problem / solution / changelog)

Summary

Fix malformed JSON aggregation in multi-tool-call streaming for the /v1/responses non-Harmony path.

Closes #39426 (sub-issue 1: tool call JSON aggregation)

What's the bug?

When a model makes multiple tool calls in a single response (e.g., "check the weather in NYC and London"), the function_call_arguments.done SSE event contains all tool calls' arguments concatenated:

{"city":"NYC"}{"city":"London"}

This is invalid JSON. The official OpenAI Python SDK crashes with an AssertionError when parsing it.

Root cause

_process_simple_streaming_events() in serving.py uses previous_delta_messages to accumulate argument deltas. This list is never reset between tool calls only on reasoningcontent transitions. At finalization, all deltas get "".join()-ed into one string.

Additionally, the tool-call transition code emits wrong event types (output_text.done, content_part.done, output_item.done with ResponseOutputMessage) instead of proper function call events.

The Harmony path is unaffected because it uses state.reset_for_new_item() between tool calls.

Fix

Minimal surgical change to the transition block:

  1. Added prev_was_tool_call check to distinguish tooltool vs texttool transitions
  2. Tooltool: emit correct function_call_arguments.done + output_item.done(ResponseFunctionToolCall) for the outgoing tool call
  3. Texttool: preserve existing behavior
  4. Reset previous_delta_messages = [] after both branches, before setting up the new tool call

Bonus fix: streaming_events.py emit_function_call_done_events() used item_id= instead of id= on ResponseFunctionToolCall. Because pydantic extra="allow", this silently created a junk field while id stayed None.

Before After

Before (2 tool calls):

output_item.added (tool 0)
  function_call_arguments.delta  N
  output_text.done               wrong event type
  content_part.done              wrong event type
  output_item.done (message)     wrong item type
output_item.added (tool 1)
  function_call_arguments.delta  N
  function_call_arguments.done (tool0 + tool1 concatenated)  corrupted
  output_item.done (function_call with concatenated args)    corrupted

After:

output_item.added (tool 0)
  function_call_arguments.delta  N
  function_call_arguments.done (only tool 0 args)  correct
  output_item.done (function_call for tool 0)      correct
output_item.added (tool 1)
  function_call_arguments.delta  N
  function_call_arguments.done (only tool 1 args)  correct
  output_item.done (function_call for tool 1)      correct

Tests added

4 unit tests in test_serving_responses.py:

TestValidates
test_tool_call_transition_emits_correct_done_eventsTooltool emits function_call_arguments.done, not output_text.done
test_multi_tool_call_arguments_not_concatenatedEach done event has its own valid JSON
test_multi_tool_call_distinct_output_indicesUnique, increasing output_index per tool call
test_single_tool_call_unchangedRegression: single tool call still works

Files changed

FileChange
vllm/entrypoints/openai/responses/serving.pyFix transition block + reset previous_delta_messages
vllm/entrypoints/openai/responses/streaming_events.pyFix item_id= id= in emit_function_call_done_events()
tests/entrypoints/openai/responses/test_serving_responses.pyAdd 4 multi-tool-call unit tests

Changed files

  • tests/entrypoints/openai/responses/test_serving_responses.py (modified, +347/-0)
  • vllm/entrypoints/openai/responses/serving.py (modified, +78/-37)
  • vllm/entrypoints/openai/responses/streaming_events.py (modified, +1/-3)

Code Example

root@spark:/workspace/vllm# python3 collect_env.py
Collecting environment information...
==============================
        System Info
==============================
OS                           : Ubuntu 24.04.4 LTS (aarch64)
GCC version                  : (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0
Clang version                : Could not collect
CMake version                : Could not collect
Libc version                 : glibc-2.39

==============================
       PyTorch Info
==============================
PyTorch version              : 2.12.0.dev20260406+cu130
Is debug build               : False
CUDA used to build PyTorch   : 13.0
ROCM used to build PyTorch   : N/A

==============================
      Python Environment
==============================
Python version               : 3.12.3 (main, Mar  3 2026, 12:15:18) [GCC 13.3.0] (64-bit runtime)
Python platform              : Linux-6.17.0-1014-nvidia-aarch64-with-glibc2.39

==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : 13.2.51
CUDA_MODULE_LOADING set to   :
GPU models and configuration : GPU 0: NVIDIA GB10
Nvidia driver version        : 580.142
cuDNN version                : Probably one of the following:
/usr/lib/aarch64-linux-gnu/libcudnn.so.9.20.0
/usr/lib/aarch64-linux-gnu/libcudnn_adv.so.9.20.0
/usr/lib/aarch64-linux-gnu/libcudnn_cnn.so.9.20.0
/usr/lib/aarch64-linux-gnu/libcudnn_engines_precompiled.so.9.20.0
/usr/lib/aarch64-linux-gnu/libcudnn_engines_runtime_compiled.so.9.20.0
/usr/lib/aarch64-linux-gnu/libcudnn_graph.so.9.20.0
/usr/lib/aarch64-linux-gnu/libcudnn_heuristic.so.9.20.0
/usr/lib/aarch64-linux-gnu/libcudnn_ops.so.9.20.0
HIP runtime version          : N/A
MIOpen runtime version       : N/A
Is XNNPACK available         : True

==============================
          CPU Info
==============================
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  20
On-line CPU(s) list:                     0-19
Vendor ID:                               ARM
Model name:                              Cortex-X925
Model:                                   1
Thread(s) per core:                      1
Core(s) per socket:                      10
Socket(s):                               1
Stepping:                                r0p1
Frequency boost:                         disabled
CPU(s) scaling MHz:                      100%
CPU max MHz:                             3900.0000
CPU min MHz:                             1378.0000
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh bti ecv afp wfxt
Model name:                              Cortex-A725
Model:                                   1
Thread(s) per core:                      1
Core(s) per socket:                      10
Socket(s):                               1
Stepping:                                r0p1
CPU(s) scaling MHz:                      100%
CPU max MHz:                             2808.0000
CPU min MHz:                             338.0000
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh bti ecv afp wfxt
L1d cache:                               1.3 MiB (20 instances)
L1i cache:                               1.3 MiB (20 instances)
L2 cache:                                25 MiB (20 instances)
L3 cache:                                24 MiB (2 instances)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-19
Vulnerability Gather data sampling:      Not affected
Vulnerability Ghostwrite:                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 Old microcode:             Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

==============================
Versions of relevant libraries
==============================
[pip3] flashinfer-python==0.6.7
[pip3] numpy==2.2.6
[pip3] nvidia-cublas==13.1.0.3
[pip3] nvidia-cuda-cupti==13.0.85
[pip3] nvidia-cuda-nvrtc==13.0.88
[pip3] nvidia-cuda-runtime==13.0.96
[pip3] nvidia-cudnn-cu13==9.20.0.48
[pip3] nvidia-cudnn-frontend==1.18.0
[pip3] nvidia-cufft==12.0.0.61
[pip3] nvidia-cufile==1.15.1.6
[pip3] nvidia-curand==10.4.0.35
[pip3] nvidia-cusolver==12.0.4.66
[pip3] nvidia-cusparse==12.6.3.3
[pip3] nvidia-cusparselt-cu13==0.8.1
[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-cu13==2.29.7
[pip3] nvidia-nvjitlink==13.0.88
[pip3] nvidia-nvshmem-cu13==3.4.5
[pip3] nvidia-nvtx==13.0.85
[pip3] pyzmq==27.1.0
[pip3] torch==2.12.0.dev20260406+cu130
[pip3] torch_c_dlpack_ext==0.1.5
[pip3] torchaudio==2.11.0.dev20260402+cu130
[pip3] torchvision==0.27.0.dev20260406+cu130
[pip3] transformers==5.5.0
[pip3] triton==3.7.0+git9c288bc5
[conda] Could not collect

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.19.1.dev0+g2a69949bd.d20260408 (git sha: 2a69949bd, date: 20260408)
vLLM Build Flags:
  CUDA Archs: 12.1a; ROCm: Disabled
GPU Topology:
        GPU0    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      0-19    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
==============================
NVIDIA_VISIBLE_DEVICES=all
NVIDIA_REQUIRE_CUDA=cuda>=13.2 brand=unknown,driver>=535,driver<536 brand=grid,driver>=535,driver<536 brand=tesla,driver>=535,driver<536 brand=nvidia,driver>=535,driver<536 brand=quadro,driver>=535,driver<536 brand=quadrortx,driver>=535,driver<536 brand=nvidiartx,driver>=535,driver<536 brand=vapps,driver>=535,driver<536 brand=vpc,driver>=535,driver<536 brand=vcs,driver>=535,driver<536 brand=vws,driver>=535,driver<536 brand=cloudgaming,driver>=535,driver<536 brand=unknown,driver>=570,driver<571 brand=grid,driver>=570,driver<571 brand=tesla,driver>=570,driver<571 brand=nvidia,driver>=570,driver<571 brand=quadro,driver>=570,driver<571 brand=quadrortx,driver>=570,driver<571 brand=nvidiartx,driver>=570,driver<571 brand=vapps,driver>=570,driver<571 brand=vpc,driver>=570,driver<571 brand=vcs,driver>=570,driver<571 brand=vws,driver>=570,driver<571 brand=cloudgaming,driver>=570,driver<571 brand=unknown,driver>=580,driver<581 brand=grid,driver>=580,driver<581 brand=tesla,driver>=580,driver<581 brand=nvidia,driver>=580,driver<581 brand=quadro,driver>=580,driver<581 brand=quadrortx,driver>=580,driver<581 brand=nvidiartx,driver>=580,driver<581 brand=vapps,driver>=580,driver<581 brand=vpc,driver>=580,driver<581 brand=vcs,driver>=580,driver<581 brand=vws,driver>=580,driver<581 brand=cloudgaming,driver>=580,driver<581 brand=unknown,driver>=590,driver<591 brand=grid,driver>=590,driver<591 brand=tesla,driver>=590,driver<591 brand=nvidia,driver>=590,driver<591 brand=quadro,driver>=590,driver<591 brand=quadrortx,driver>=590,driver<591 brand=nvidiartx,driver>=590,driver<591 brand=vapps,driver>=590,driver<591 brand=vpc,driver>=590,driver<591 brand=vcs,driver>=590,driver<591 brand=vws,driver>=590,driver<591 brand=cloudgaming,driver>=590,driver<591
TORCH_CUDA_ARCH_LIST=12.1a
NVIDIA_DRIVER_CAPABILITIES=compute,utility
NVIDIA_PRODUCT_NAME=CUDA
CUDA_VERSION=13.2.0
MAX_JOBS=16
LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib64
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
TORCHINDUCTOR_CACHE_DIR=/tmp/torchinductor_root
RAW_BUFFERClick to expand / collapse

Your current environment

compile_project_v2.py

<details> <summary>The output of <code>python collect_env.py</code></summary> 
root@spark:/workspace/vllm# python3 collect_env.py
Collecting environment information...
==============================
        System Info
==============================
OS                           : Ubuntu 24.04.4 LTS (aarch64)
GCC version                  : (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0
Clang version                : Could not collect
CMake version                : Could not collect
Libc version                 : glibc-2.39

==============================
       PyTorch Info
==============================
PyTorch version              : 2.12.0.dev20260406+cu130
Is debug build               : False
CUDA used to build PyTorch   : 13.0
ROCM used to build PyTorch   : N/A

==============================
      Python Environment
==============================
Python version               : 3.12.3 (main, Mar  3 2026, 12:15:18) [GCC 13.3.0] (64-bit runtime)
Python platform              : Linux-6.17.0-1014-nvidia-aarch64-with-glibc2.39

==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : 13.2.51
CUDA_MODULE_LOADING set to   :
GPU models and configuration : GPU 0: NVIDIA GB10
Nvidia driver version        : 580.142
cuDNN version                : Probably one of the following:
/usr/lib/aarch64-linux-gnu/libcudnn.so.9.20.0
/usr/lib/aarch64-linux-gnu/libcudnn_adv.so.9.20.0
/usr/lib/aarch64-linux-gnu/libcudnn_cnn.so.9.20.0
/usr/lib/aarch64-linux-gnu/libcudnn_engines_precompiled.so.9.20.0
/usr/lib/aarch64-linux-gnu/libcudnn_engines_runtime_compiled.so.9.20.0
/usr/lib/aarch64-linux-gnu/libcudnn_graph.so.9.20.0
/usr/lib/aarch64-linux-gnu/libcudnn_heuristic.so.9.20.0
/usr/lib/aarch64-linux-gnu/libcudnn_ops.so.9.20.0
HIP runtime version          : N/A
MIOpen runtime version       : N/A
Is XNNPACK available         : True

==============================
          CPU Info
==============================
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  20
On-line CPU(s) list:                     0-19
Vendor ID:                               ARM
Model name:                              Cortex-X925
Model:                                   1
Thread(s) per core:                      1
Core(s) per socket:                      10
Socket(s):                               1
Stepping:                                r0p1
Frequency boost:                         disabled
CPU(s) scaling MHz:                      100%
CPU max MHz:                             3900.0000
CPU min MHz:                             1378.0000
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh bti ecv afp wfxt
Model name:                              Cortex-A725
Model:                                   1
Thread(s) per core:                      1
Core(s) per socket:                      10
Socket(s):                               1
Stepping:                                r0p1
CPU(s) scaling MHz:                      100%
CPU max MHz:                             2808.0000
CPU min MHz:                             338.0000
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh bti ecv afp wfxt
L1d cache:                               1.3 MiB (20 instances)
L1i cache:                               1.3 MiB (20 instances)
L2 cache:                                25 MiB (20 instances)
L3 cache:                                24 MiB (2 instances)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-19
Vulnerability Gather data sampling:      Not affected
Vulnerability Ghostwrite:                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 Old microcode:             Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

==============================
Versions of relevant libraries
==============================
[pip3] flashinfer-python==0.6.7
[pip3] numpy==2.2.6
[pip3] nvidia-cublas==13.1.0.3
[pip3] nvidia-cuda-cupti==13.0.85
[pip3] nvidia-cuda-nvrtc==13.0.88
[pip3] nvidia-cuda-runtime==13.0.96
[pip3] nvidia-cudnn-cu13==9.20.0.48
[pip3] nvidia-cudnn-frontend==1.18.0
[pip3] nvidia-cufft==12.0.0.61
[pip3] nvidia-cufile==1.15.1.6
[pip3] nvidia-curand==10.4.0.35
[pip3] nvidia-cusolver==12.0.4.66
[pip3] nvidia-cusparse==12.6.3.3
[pip3] nvidia-cusparselt-cu13==0.8.1
[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-cu13==2.29.7
[pip3] nvidia-nvjitlink==13.0.88
[pip3] nvidia-nvshmem-cu13==3.4.5
[pip3] nvidia-nvtx==13.0.85
[pip3] pyzmq==27.1.0
[pip3] torch==2.12.0.dev20260406+cu130
[pip3] torch_c_dlpack_ext==0.1.5
[pip3] torchaudio==2.11.0.dev20260402+cu130
[pip3] torchvision==0.27.0.dev20260406+cu130
[pip3] transformers==5.5.0
[pip3] triton==3.7.0+git9c288bc5
[conda] Could not collect

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.19.1.dev0+g2a69949bd.d20260408 (git sha: 2a69949bd, date: 20260408)
vLLM Build Flags:
  CUDA Archs: 12.1a; ROCm: Disabled
GPU Topology:
        GPU0    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      0-19    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
==============================
NVIDIA_VISIBLE_DEVICES=all
NVIDIA_REQUIRE_CUDA=cuda>=13.2 brand=unknown,driver>=535,driver<536 brand=grid,driver>=535,driver<536 brand=tesla,driver>=535,driver<536 brand=nvidia,driver>=535,driver<536 brand=quadro,driver>=535,driver<536 brand=quadrortx,driver>=535,driver<536 brand=nvidiartx,driver>=535,driver<536 brand=vapps,driver>=535,driver<536 brand=vpc,driver>=535,driver<536 brand=vcs,driver>=535,driver<536 brand=vws,driver>=535,driver<536 brand=cloudgaming,driver>=535,driver<536 brand=unknown,driver>=570,driver<571 brand=grid,driver>=570,driver<571 brand=tesla,driver>=570,driver<571 brand=nvidia,driver>=570,driver<571 brand=quadro,driver>=570,driver<571 brand=quadrortx,driver>=570,driver<571 brand=nvidiartx,driver>=570,driver<571 brand=vapps,driver>=570,driver<571 brand=vpc,driver>=570,driver<571 brand=vcs,driver>=570,driver<571 brand=vws,driver>=570,driver<571 brand=cloudgaming,driver>=570,driver<571 brand=unknown,driver>=580,driver<581 brand=grid,driver>=580,driver<581 brand=tesla,driver>=580,driver<581 brand=nvidia,driver>=580,driver<581 brand=quadro,driver>=580,driver<581 brand=quadrortx,driver>=580,driver<581 brand=nvidiartx,driver>=580,driver<581 brand=vapps,driver>=580,driver<581 brand=vpc,driver>=580,driver<581 brand=vcs,driver>=580,driver<581 brand=vws,driver>=580,driver<581 brand=cloudgaming,driver>=580,driver<581 brand=unknown,driver>=590,driver<591 brand=grid,driver>=590,driver<591 brand=tesla,driver>=590,driver<591 brand=nvidia,driver>=590,driver<591 brand=quadro,driver>=590,driver<591 brand=quadrortx,driver>=590,driver<591 brand=nvidiartx,driver>=590,driver<591 brand=vapps,driver>=590,driver<591 brand=vpc,driver>=590,driver<591 brand=vcs,driver>=590,driver<591 brand=vws,driver>=590,driver<591 brand=cloudgaming,driver>=590,driver<591
TORCH_CUDA_ARCH_LIST=12.1a
NVIDIA_DRIVER_CAPABILITIES=compute,utility
NVIDIA_PRODUCT_NAME=CUDA
CUDA_VERSION=13.2.0
MAX_JOBS=16
LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib64
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
TORCHINDUCTOR_CACHE_DIR=/tmp/torchinductor_root
</details>

🐛 Describe the bug

Here’s what appears to be some legitimate vLLM errors. I have a custom AI client using the Python openai library. Gemini 3 Flash (Preview) generated this after working with me on my client implementation. The context for finding this is that we are trying to make my client work with the official OpenAI library and vLLM’s /v1/responses endpoint. We currently have to use a lower-level manual SSE parser to handle the stream because vLLM’s output is incompatible with the SDK’s internal assertions.

Bug Report: vLLM /v1/responses Protocol Violations Breaking Official OpenAI SDK Environment:

  1. CRITICAL: Malformed JSON Aggregation in Tool Calls Issue: In turns requiring multiple simultaneous tool calls, vLLM incorrectly concatenates JSON argument strings in the .done event instead of scoping them to their specific output_index. This causes the official openai-python SDK to crash with an AssertionError.

Reproduction Proof (Raw SSE Trace):

Tool Call 1 started (output_index: 2) Tool Call 2 started (output_index: 3) Raw arguments field content in .done event: {"file_path": "foo.txt", "reason": "..."}{"file_path": "bar.txt", "reason": "..."} Violation: As per Table E of the OpenAI Field Guide, the arguments field in a function_call_arguments.done event must be a single, valid JSON object. vLLM is producing a }{ concatenation, which is invalid JSON.

Impact: The official SDK’s high-level parser (client.responses.stream()) validates the JSON buffer during assembly. Encountering concatenated objects causes an internal state mismatch and an immediate crash.

  1. PERFORMANCE: Redundant Instruction Echoing (Token Waste) Issue: vLLM echoes the full instructions block (system prompt) and all request parameters in multiple SSE events at the start of every stream.

Reproduction Proof:

  • Instruction Size: 1,111 characters
  • Event 1 (response.created): Instructions echoed back (1,111 chars)
  • Event 2 (response.in_progress): Instructions echoed back again (1,111 chars)
  • Total Redundant Characters: 2,222 (Double the prompt size).

The Impact: This violates the efficiency principles of the Responses API. In applications with large system prompts, this double-echoing induces significant network latency and bandwidth waste. Standard behavior should only echo parameters once (typically in response.created).

  1. COMPATIBILITY: Non-Standard Reasoning Lifecycle Events Issue: vLLM uses custom event types response.reasoning_part.added and response.reasoning_part.done.

Violation: While reasoning_text.delta is a recognized standard, the lifecycle events for these parts should follow the standard response.content_part.added (type: reasoning_text) schema defined in the OpenAI Field Guide (Table D).

Impact: Invented event names like reasoning_part cause drift from the official SDK’s event-dispatch registry and trigger internal assertions during stream parsing.

Suggested Remediation: Fix Aggregator: Ensure function_call_arguments strings contain only the JSON for the specific output_index being finalized. Do not concatenate.

Align SSE Schema: Transition reasoning_part lifecycle events to the standard content_part schema.

Optimize Network Traffic: Suppress the redundant instructions echo in events following the initial response.created.

Attached is a script to reproduce these problems. Please let me know if you need more information.

reproduce_vllm_bugs.py

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 most likely fix involves updating the vLLM protocol implementation to adhere to the OpenAI Responses API standards, specifically addressing malformed JSON aggregation, redundant instruction echoing, and non-standard reasoning lifecycle events.

Guidance

  1. Fix JSON Aggregation: Ensure that function_call_arguments strings in the .done event are scoped to their specific output_index and do not concatenate JSON objects.
  2. Optimize SSE Events: Suppress redundant instructions echoing in events following the initial response.created to reduce network latency and bandwidth waste.
  3. Standardize Reasoning Lifecycle Events: Transition reasoning_part lifecycle events to the standard content_part schema to align with the OpenAI Field Guide and prevent internal assertions during stream parsing.

Example

To illustrate the fix for malformed JSON aggregation, consider the following example:

# Incorrect (concatenated JSON objects)
{"file_path": "foo.txt", "reason": "..."}{"file_path": "bar.txt", "reason": "..."}

# Correct (separate JSON objects for each output_index)
[
  {"output_index": 2, "arguments": {"file_path": "foo.txt", "reason": "..."}},
  {"output_index": 3, "arguments": {"file_path": "bar.txt", "reason": "..."}}
]

Notes

The provided reproduction script (reproduce_vllm_bugs.py) can be used to verify the fixes for these issues.

Recommendation

Apply the suggested remediation steps to update the vLLM protocol implementation and ensure compatibility with the official OpenAI SDK.

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…
#api#environment variable#retrieval issue#search optimization#API routing

Still need to ship something?

×6

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

Back to top recommendations

TRENDING