- With `logprobs_mode=raw_logits`, the returned `logprob` values are pre-softmax raw logits (unbounded; typically include positive values for the sampled token). - With `logprob_token_ids=[100, 1000]`, those two vocab ids appear in the response regardless of where they rank in the natural top-k.

vllm - 💡(How to fix) Fix [Bug]: OpenAI chat/completion endpoints ignore --logprobs-mode and don't accept logprob_token_ids

vllm2026-05-23 01:46:22

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.

GitHub issue URL

Helpful · Quick feedback

The OpenAI-compat HTTP endpoints (/v1/chat/completions and /v1/completions) don't honor --logprobs-mode per request, and don't accept logprob_token_ids. Both are supported on the Python LLM class, and logprob_token_ids is accepted on /generative_scoring, but the standard OpenAI-compat routes that most clients use have no path to either.

Concretely:

The chat/completion endpoint always returns log_softmax(logit) as the logprob field, regardless of how the server was launched. A user who needs raw logits per request has no path short of restarting the server with --logprobs-mode raw_logits, which then affects every client globally.
logprob_token_ids is only accepted on /generative_scoring, not on /v1/chat/completions / /v1/completions.

Error Message

The server boots cleanly and there is no error logged. The fields are simply not parsed off the request body.

Root Cause

The sampler-side support already exists upstream. Sampler.forward honors logprobs_mode_override on both the top-k logprobs path and the labeled-vocab gather path, and SamplingMetadata.logprob_token_ids is consumed correctly. The gap is purely in the OpenAI-compat HTTP layer: the request fields aren't added to ChatCompletionRequest / CompletionRequest, so they never reach SamplingParams.

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): 48 On-line CPU(s) list: 0-47 Vendor ID: AuthenticAMD Model name: AMD EPYC 9B45 CPU family: 26 Model: 2 Thread(s) per core: 2 Core(s) per socket: 24 Socket(s): 1 Stepping: 1 BogoMIPS: 5399.99 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 tsc_known_freq pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm cmp_legacy cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw topoext mwaitx invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced vmmcall fsgsbase tsc_adjust bmi1 avx2 smep bmi2 invpcid avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves avx_vnni avx512_bf16 clzero xsaveerptr wbnoinvd arat avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq rdpid movdiri movdir64b avx512_vp2intersect flush_l1d Hypervisor vendor: KVM Virtualization type: full L1d cache: 1.1 MiB (24 instances) L1i cache: 768 KiB (24 instances) L2 cache: 24 MiB (24 instances) L3 cache: 96 MiB (3 instances) NUMA node(s): 1 NUMA node0 CPU(s): 0-47 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: Not affected 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: Not affected Vulnerability Tsx async abort: Not affected Vulnerability Vmscape: Not affected

Code Example

Collecting environment information...
==============================
        System Info
==============================
OS                           : Debian GNU/Linux 12 (bookworm) (x86_64)
GCC version                  : (Debian 12.2.0-14+deb12u1) 12.2.0
Clang version                : Could not collect
CMake version                : Could not collect
Libc version                 : glibc-2.36

==============================
       PyTorch Info
==============================
PyTorch version              : 2.11.0+cu130
Is debug build               : False
CUDA used to build PyTorch   : 13.0
ROCM used to build PyTorch   : N/A
XPU used to build PyTorch    : N/A

==============================
      Python Environment
==============================
Python version               : 3.12.9 (main, Apr 28 2026, 03:56:56) [GCC 12.2.0] (64-bit runtime)
Python platform              : Linux-6.1.0-45-cloud-amd64-x86_64-with-glibc2.36
    
==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : 13.2.78
CUDA_MODULE_LOADING set to   : 
GPU models and configuration : GPU 0: NVIDIA RTX PRO 6000 Blackwell Server Edition
Nvidia driver version        : 595.58.03
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):                                  48
On-line CPU(s) list:                     0-47
Vendor ID:                               AuthenticAMD
Model name:                              AMD EPYC 9B45
CPU family:                              26
Model:                                   2
Thread(s) per core:                      2
Core(s) per socket:                      24
Socket(s):                               1
Stepping:                                1
BogoMIPS:                                5399.99
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 tsc_known_freq pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm cmp_legacy cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw topoext mwaitx invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced vmmcall fsgsbase tsc_adjust bmi1 avx2 smep bmi2 invpcid avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves avx_vnni avx512_bf16 clzero xsaveerptr wbnoinvd arat avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq rdpid movdiri movdir64b avx512_vp2intersect flush_l1d
Hypervisor vendor:                       KVM
Virtualization type:                     full
L1d cache:                               1.1 MiB (24 instances)
L1i cache:                               768 KiB (24 instances)
L2 cache:                                24 MiB (24 instances)
L3 cache:                                96 MiB (3 instances)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-47
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:      Not affected
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:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

==============================
Versions of relevant libraries
==============================
[pip3] flashinfer-python==0.6.11.post2
[pip3] numpy==2.3.5
[pip3] nvidia-cublas==13.1.0.3
[pip3] nvidia-cuda-cccl==13.2.75
[pip3] nvidia-cuda-crt==13.2.78
[pip3] nvidia-cuda-cupti==13.0.85
[pip3] nvidia-cuda-nvcc==13.2.78
[pip3] nvidia-cuda-nvrtc==13.0.88
[pip3] nvidia-cuda-runtime==13.0.96
[pip3] nvidia-cudnn-cu13==9.19.0.56
[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==0.8.0
[pip3] nvidia-cutlass-dsl==4.5.0
[pip3] nvidia-cutlass-dsl-libs-base==4.5.0
[pip3] nvidia-cutlass-dsl-libs-cu13==4.5.0
[pip3] nvidia-ml-py==13.595.45
[pip3] nvidia-nccl-cu13==2.28.9
[pip3] nvidia-nvjitlink==13.0.88
[pip3] nvidia-nvshmem-cu13==3.4.5
[pip3] nvidia-nvtx==13.0.85
[pip3] nvidia-nvvm==13.2.78
[pip3] pyzmq==27.1.0
[pip3] tokenspeed-triton==3.7.10.post20260505
[pip3] torch==2.11.0
[pip3] torch_c_dlpack_ext==0.1.5
[pip3] torchaudio==2.11.0
[pip3] torchvision==0.26.0
[pip3] transformers==5.6.2
[pip3] triton==3.6.0
[conda] Could not collect

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.21.1rc1.dev236+g237288600 (git sha: 237288600)
vLLM Build Flags:
  CUDA Archs: Not Set; ROCm: Disabled; XPU: Disabled
GPU Topology:
  	[4mGPU0	CPU Affinity	NUMA Affinity	GPU NUMA ID[0m
GPU0	 X 	0-47	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
==============================
LD_LIBRARY_PATH=/usr/local/cuda-13.2/lib64:
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
TORCHINDUCTOR_CACHE_DIR=/tmp/torchinductor_langzhao

---

vllm serve Qwen/Qwen2.5-1.5B-Instruct --max-logprobs -1

---

from openai import OpenAI

client = OpenAI(base_url="http://localhost:8000/v1", api_key="dummy")

resp = client.chat.completions.create(
    model="Qwen/Qwen2.5-1.5B-Instruct",
    messages=[{"role": "user", "content": "Hello"}],
    max_tokens=1,
    temperature=0,
    logprobs=True,
    top_logprobs=5,
    extra_body={
        "logprobs_mode": "raw_logits",      # silently ignored
        "logprob_token_ids": [100, 1000],   # silently ignored
    },
)

for e in resp.choices[0].logprobs.content[0].top_logprobs:
    print(e.token, e.logprob)

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                           : Debian GNU/Linux 12 (bookworm) (x86_64)
GCC version                  : (Debian 12.2.0-14+deb12u1) 12.2.0
Clang version                : Could not collect
CMake version                : Could not collect
Libc version                 : glibc-2.36

==============================
       PyTorch Info
==============================
PyTorch version              : 2.11.0+cu130
Is debug build               : False
CUDA used to build PyTorch   : 13.0
ROCM used to build PyTorch   : N/A
XPU used to build PyTorch    : N/A

==============================
      Python Environment
==============================
Python version               : 3.12.9 (main, Apr 28 2026, 03:56:56) [GCC 12.2.0] (64-bit runtime)
Python platform              : Linux-6.1.0-45-cloud-amd64-x86_64-with-glibc2.36
    
==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : 13.2.78
CUDA_MODULE_LOADING set to   : 
GPU models and configuration : GPU 0: NVIDIA RTX PRO 6000 Blackwell Server Edition
Nvidia driver version        : 595.58.03
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):                                  48
On-line CPU(s) list:                     0-47
Vendor ID:                               AuthenticAMD
Model name:                              AMD EPYC 9B45
CPU family:                              26
Model:                                   2
Thread(s) per core:                      2
Core(s) per socket:                      24
Socket(s):                               1
Stepping:                                1
BogoMIPS:                                5399.99
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 tsc_known_freq pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm cmp_legacy cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw topoext mwaitx invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced vmmcall fsgsbase tsc_adjust bmi1 avx2 smep bmi2 invpcid avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves avx_vnni avx512_bf16 clzero xsaveerptr wbnoinvd arat avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq rdpid movdiri movdir64b avx512_vp2intersect flush_l1d
Hypervisor vendor:                       KVM
Virtualization type:                     full
L1d cache:                               1.1 MiB (24 instances)
L1i cache:                               768 KiB (24 instances)
L2 cache:                                24 MiB (24 instances)
L3 cache:                                96 MiB (3 instances)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-47
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:      Not affected
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:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

==============================
Versions of relevant libraries
==============================
[pip3] flashinfer-python==0.6.11.post2
[pip3] numpy==2.3.5
[pip3] nvidia-cublas==13.1.0.3
[pip3] nvidia-cuda-cccl==13.2.75
[pip3] nvidia-cuda-crt==13.2.78
[pip3] nvidia-cuda-cupti==13.0.85
[pip3] nvidia-cuda-nvcc==13.2.78
[pip3] nvidia-cuda-nvrtc==13.0.88
[pip3] nvidia-cuda-runtime==13.0.96
[pip3] nvidia-cudnn-cu13==9.19.0.56
[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==0.8.0
[pip3] nvidia-cutlass-dsl==4.5.0
[pip3] nvidia-cutlass-dsl-libs-base==4.5.0
[pip3] nvidia-cutlass-dsl-libs-cu13==4.5.0
[pip3] nvidia-ml-py==13.595.45
[pip3] nvidia-nccl-cu13==2.28.9
[pip3] nvidia-nvjitlink==13.0.88
[pip3] nvidia-nvshmem-cu13==3.4.5
[pip3] nvidia-nvtx==13.0.85
[pip3] nvidia-nvvm==13.2.78
[pip3] pyzmq==27.1.0
[pip3] tokenspeed-triton==3.7.10.post20260505
[pip3] torch==2.11.0
[pip3] torch_c_dlpack_ext==0.1.5
[pip3] torchaudio==2.11.0
[pip3] torchvision==0.26.0
[pip3] transformers==5.6.2
[pip3] triton==3.6.0
[conda] Could not collect

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.21.1rc1.dev236+g237288600 (git sha: 237288600)
vLLM Build Flags:
  CUDA Archs: Not Set; ROCm: Disabled; XPU: Disabled
GPU Topology:
  	[4mGPU0	CPU Affinity	NUMA Affinity	GPU NUMA ID[0m
GPU0	 X 	0-47	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
==============================
LD_LIBRARY_PATH=/usr/local/cuda-13.2/lib64:
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
TORCHINDUCTOR_CACHE_DIR=/tmp/torchinductor_langzhao

</details>

🐛 Describe the bug

Summary

Concretely:

The chat/completion endpoint always returns log_softmax(logit) as the logprob field, regardless of how the server was launched. A user who needs raw logits per request has no path short of restarting the server with --logprobs-mode raw_logits, which then affects every client globally.
logprob_token_ids is only accepted on /generative_scoring, not on /v1/chat/completions / /v1/completions.

Why this matters

Multilabel postprocessors score sigmoid(raw_logit[label_token_id]) independently per label, which requires two things the HTTP endpoint doesn't provide:

Raw logit values at specific vocab ids (not post-softmax). The current logprob field is post-softmax log_softmax(logit); applying sigmoid to it gives systematically wrong per-class probabilities.
The ability to retrieve logits at a fixed list of label vocab ids, regardless of where they rank in the model's natural top-k distribution. With only top_logprobs=N, label ids outside the top-N are silently dropped.

Steps to reproduce

Start a server on the default mode (no --logprobs-mode flag):

vllm serve Qwen/Qwen2.5-1.5B-Instruct --max-logprobs -1

Send a request that asks for raw logits and a specific vocab id:

from openai import OpenAI

client = OpenAI(base_url="http://localhost:8000/v1", api_key="dummy")

resp = client.chat.completions.create(
    model="Qwen/Qwen2.5-1.5B-Instruct",
    messages=[{"role": "user", "content": "Hello"}],
    max_tokens=1,
    temperature=0,
    logprobs=True,
    top_logprobs=5,
    extra_body={
        "logprobs_mode": "raw_logits",      # silently ignored
        "logprob_token_ids": [100, 1000],   # silently ignored
    },
)

for e in resp.choices[0].logprobs.content[0].top_logprobs:
    print(e.token, e.logprob)

Expected behavior

With logprobs_mode=raw_logits, the returned logprob values are pre-softmax raw logits (unbounded; typically include positive values for the sampled token).
With logprob_token_ids=[100, 1000], those two vocab ids appear in the response regardless of where they rank in the natural top-k.

Actual behavior

Both extension fields are silently dropped. All returned logprob values are <= 0 (post-softmax log_softmax). The response contains only the natural top-5 most-likely tokens; vocab ids 100 and 1000 are absent unless they happen to be in the top-5 by likelihood.

The server boots cleanly and there is no error logged. The fields are simply not parsed off the request body.

Root cause

Proposed fix

#43463 adds the two fields to ChatCompletionRequest and CompletionRequest and plumbs them through SamplingParams → InputBatch.logprobs_mode → SamplingMetadata.logprobs_mode_override → Sampler.forward(..., logprobs_mode_override=...). Includes unit tests for batch-level mode aggregation and end-to-end HTTP tests for the four logprobs_mode values.

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.

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?

FAQ

Expected behavior

With logprobs_mode=raw_logits, the returned logprob values are pre-softmax raw logits (unbounded; typically include positive values for the sampled token).
With logprob_token_ids=[100, 1000], those two vocab ids appear in the response regardless of where they rank in the natural top-k.

Still need to ship something?

×6

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

Data

Security

Network

Code

UI/UX

Text

System

Multimedia

Protocol

API

Engineering

vllm - 💡(How to fix) Fix [Bug]: OpenAI chat/completion endpoints ignore --logprobs-mode and don't accept logprob_token_ids

Recommended Tools

GitHub issue graph ai analysis

Error Message

Root Cause

Fix Action

Fix / Workaround

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

Code Example

Your current environment

🐛 Describe the bug

Summary

Why this matters

Steps to reproduce

Expected behavior

Actual behavior

Root cause

Proposed fix

Before submitting a new issue...

FAQ

Expected behavior

Still need to ship something?

TRENDING