[Bug]: 'DeepseekV2Model' object has no attribute 'config' when enabling P/D Disaggregation

[JewelRoam]

Your current environment

The output of `python collect_env.py`

INFO 03-05 19:23:48 __init__.py:190] Automatically detected platform cuda.
Collecting environment information...
PyTorch version: 2.5.1+cu124
Is debug build: False
CUDA used to build PyTorch: 12.4
ROCM used to build PyTorch: N/A

OS: Ubuntu 20.04.6 LTS (x86_64)
GCC version: (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0
Clang version: Could not collect
CMake version: Could not collect
Libc version: glibc-2.31

Python version: 3.10.16 (main, Dec  4 2024, 08:53:37) [GCC 9.4.0] (64-bit runtime)
Python platform: Linux-5.4.0-205-generic-x86_64-with-glibc2.31
Is CUDA available: True
CUDA runtime version: 12.5.82
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: 
GPU 0: NVIDIA H800 PCIe
GPU 1: NVIDIA H800 PCIe
GPU 2: NVIDIA H800 PCIe
GPU 3: NVIDIA H800 PCIe

Nvidia driver version: 550.90.07
cuDNN version: Could not collect
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Architecture:                       x86_64
CPU op-mode(s):                     32-bit, 64-bit
Byte Order:                         Little Endian
Address sizes:                      52 bits physical, 57 bits virtual
CPU(s):                             192
On-line CPU(s) list:                0-191
Thread(s) per core:                 2
Core(s) per socket:                 48
Socket(s):                          2
NUMA node(s):                       2
Vendor ID:                          GenuineIntel
CPU family:                         6
Model:                              143
Model name:                         Intel(R) Xeon(R) Platinum 8468V
Stepping:                           8
CPU MHz:                            800.016
CPU max MHz:                        3800.0000
CPU min MHz:                        800.0000
BogoMIPS:                           4800.00
Virtualization:                     VT-x
L1d cache:                          4.5 MiB
L1i cache:                          3 MiB
L2 cache:                           192 MiB
L3 cache:                           195 MiB
NUMA node0 CPU(s):                  0-47,96-143
NUMA node1 CPU(s):                  48-95,144-191
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 Retbleed:             Not affected
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; Enhanced / Automatic IBRS; IBPB conditional; RSB filling; PBRSB-eIBRS SW sequence; BHI BHI_DIS_S
Vulnerability Srbds:                Not affected
Vulnerability Tsx async abort:      Not affected
Flags:                              fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cat_l2 cdp_l3 invpcid_single cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local avx512_bf16 wbnoinvd dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req avx512vbmi umip pku ospke waitpkg avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme avx512_vpopcntdq rdpid cldemote movdiri movdir64b md_clear pconfig flush_l1d arch_capabilities

Versions of relevant libraries:
[pip3] flashinfer-python==0.2.0.post2+cu124torch2.5
[pip3] mypy-extensions==1.0.0
[pip3] numpy==1.26.4
[pip3] nvidia-cublas-cu12==12.4.5.8
[pip3] nvidia-cuda-cupti-cu12==12.4.127
[pip3] nvidia-cuda-nvrtc-cu12==12.4.127
[pip3] nvidia-cuda-runtime-cu12==12.4.127
[pip3] nvidia-cudnn-cu12==9.1.0.70
[pip3] nvidia-cufft-cu12==11.2.1.3
[pip3] nvidia-curand-cu12==10.3.5.147
[pip3] nvidia-cusolver-cu12==11.6.1.9
[pip3] nvidia-cusparse-cu12==12.3.1.170
[pip3] nvidia-cusparselt-cu12==0.6.2
[pip3] nvidia-ml-py==12.570.86
[pip3] nvidia-nccl-cu12==2.21.5
[pip3] nvidia-nvjitlink-cu12==12.4.127
[pip3] nvidia-nvtx-cu12==12.4.127
[pip3] pyzmq==26.2.1
[pip3] torch==2.5.1
[pip3] torchao==0.8.0
[pip3] torchaudio==2.5.1
[pip3] torchvision==0.20.1
[pip3] transformers==4.48.2
[pip3] triton==3.1.0
[conda] Could not collect
ROCM Version: Could not collect
Neuron SDK Version: N/A
vLLM Version: 0.7.2
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0    GPU1    GPU2    GPU3    NIC0    NIC1    NIC2    NIC3    NIC4    NIC5    NIC6    NIC7    NIC8    NIC9    NIC10   NIC11   CPU Affinity    NUMA Affinity  GPU NUMA ID
GPU0     X      PIX     SYS     SYS     PIX     PIX     PIX     PIX     PIX     NODE    NODE    SYS     SYS     SYS     SYS     SYS     0-47,96-143     0      N/A
GPU1    PIX      X      SYS     SYS     PIX     PIX     PIX     PIX     PIX     NODE    NODE    SYS     SYS     SYS     SYS     SYS     0-47,96-143     0      N/A
GPU2    SYS     SYS      X      PIX     SYS     SYS     SYS     SYS     SYS     SYS     SYS     PIX     PIX     PIX     PIX     PIX     48-95,144-191   1      N/A
GPU3    SYS     SYS     PIX      X      SYS     SYS     SYS     SYS     SYS     SYS     SYS     PIX     PIX     PIX     PIX     PIX     48-95,144-191   1      N/A
NIC0    PIX     PIX     SYS     SYS      X      PIX     PIX     PIX     PIX     NODE    NODE    SYS     SYS     SYS     SYS     SYS
NIC1    PIX     PIX     SYS     SYS     PIX      X      PIX     PIX     PIX     NODE    NODE    SYS     SYS     SYS     SYS     SYS
NIC2    PIX     PIX     SYS     SYS     PIX     PIX      X      PIX     PIX     NODE    NODE    SYS     SYS     SYS     SYS     SYS
NIC3    PIX     PIX     SYS     SYS     PIX     PIX     PIX      X      PIX     NODE    NODE    SYS     SYS     SYS     SYS     SYS
NIC4    PIX     PIX     SYS     SYS     PIX     PIX     PIX     PIX      X      NODE    NODE    SYS     SYS     SYS     SYS     SYS
NIC5    NODE    NODE    SYS     SYS     NODE    NODE    NODE    NODE    NODE     X      PIX     SYS     SYS     SYS     SYS     SYS
NIC6    NODE    NODE    SYS     SYS     NODE    NODE    NODE    NODE    NODE    PIX      X      SYS     SYS     SYS     SYS     SYS
NIC7    SYS     SYS     PIX     PIX     SYS     SYS     SYS     SYS     SYS     SYS     SYS      X      PIX     PIX     PIX     PIX
NIC8    SYS     SYS     PIX     PIX     SYS     SYS     SYS     SYS     SYS     SYS     SYS     PIX      X      PIX     PIX     PIX
NIC9    SYS     SYS     PIX     PIX     SYS     SYS     SYS     SYS     SYS     SYS     SYS     PIX     PIX      X      PIX     PIX
NIC10   SYS     SYS     PIX     PIX     SYS     SYS     SYS     SYS     SYS     SYS     SYS     PIX     PIX     PIX      X      PIX
NIC11   SYS     SYS     PIX     PIX     SYS     SYS     SYS     SYS     SYS     SYS     SYS     PIX     PIX     PIX     PIX      X 

Legend:

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

NIC Legend:

  NIC0: mlx5_0
  NIC1: mlx5_1
  NIC2: mlx5_2
  NIC3: mlx5_3
  NIC4: mlx5_4
  NIC5: mlx5_5
  NIC6: mlx5_6
  NIC7: mlx5_7
  NIC8: mlx5_8
  NIC9: mlx5_9
  NIC10: mlx5_10
  NIC11: mlx5_11

NVIDIA_VISIBLE_DEVICES=all
NCCL_IB_TC=120
NVIDIA_REQUIRE_CUDA=cuda>=12.5 brand=unknown,driver>=470,driver<471 brand=grid,driver>=470,driver<471 brand=tesla,driver>=470,driver<471 brand=nvidia,driver>=470,driver<471 brand=quadro,driver>=470,driver<471 brand=quadrortx,driver>=470,driver<471 brand=nvidiartx,driver>=470,driver<471 brand=vapps,driver>=470,driver<471 brand=vpc,driver>=470,driver<471 brand=vcs,driver>=470,driver<471 brand=vws,driver>=470,driver<471 brand=cloudgaming,driver>=470,driver<471 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>=550,driver<551 brand=grid,driver>=550,driver<551 brand=tesla,driver>=550,driver<551 brand=nvidia,driver>=550,driver<551 brand=quadro,driver>=550,driver<551 brand=quadrortx,driver>=550,driver<551 brand=nvidiartx,driver>=550,driver<551 brand=vapps,driver>=550,driver<551 brand=vpc,driver>=550,driver<551 brand=vcs,driver>=550,driver<551 brand=vws,driver>=550,driver<551 brand=cloudgaming,driver>=550,driver<551
NCCL_VERSION=2.22.3-1
NCCL_SOCKET_IFNAME=ens103f0np0
NVIDIA_DRIVER_CAPABILITIES=compute,utility
NCCL_IB_HCA=mlx5_0:1
NVIDIA_PRODUCT_NAME=CUDA
NCCL_IB_GID_INDEX=3
CUDA_VERSION=12.5.1
CUDA_VISIBLE_DEVICES=0,1,2,3
CUDA_VISIBLE_DEVICES=0,1,2,3
CUDA_DEVICE_MAX_CONNECTIONS=1
LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64
VLLM_HOST_IP=10.1.10.232
NCCL_IB_ADAPTIVE_ROUTING=1
NCCL_CUMEM_ENABLE=0
TORCHINDUCTOR_COMPILE_THREADS=1
CUDA_MODULE_LOADING=LAZY

🐛 Describe the bug

When I was trying Disaggregated Prefilling with DeepSeek-V2-Lite, I firstly started two backend with a proxy:

CUDA_VISIBLE_DEVICES=0 python3 \
    -m vllm.entrypoints.openai.api_server \
    --model /workspace/deepseek/zer/DeepSeek-V2-Lite/  \
    --port 8100 \
    --max-model-len 10000 \
    --gpu-memory-utilization 0.95 \
    --trust_remote_code \
    --kv-transfer-config \
    '{"kv_connector":"PyNcclConnector","kv_role":"kv_producer","kv_rank":0,"kv_parallel_size":2,"kv_buffer_size":5e9}' &
  CUDA_VISIBLE_DEVICES=1 python3 \
    -m vllm.entrypoints.openai.api_server \
    --model /workspace/deepseek/zer/DeepSeek-V2-Lite/  \
    --port 8200 \
    --max-model-len 10000 \
    --gpu-memory-utilization 0.95 \
    --trust_remote_code \
    --kv-transfer-config \
    '{"kv_connector":"PyNcclConnector","kv_role":"kv_consumer","kv_rank":1,"kv_parallel_size":2,"kv_buffer_size":5e9}' &
  wait_for_server 8100
  wait_for_server 8200
  python3 disagg_prefill_proxy_server.py &  # Listening on port 8000

Then I run a benchmark test:

python3 ../benchmark_serving.py \
          --backend vllm \
          --model /workspace/deepseek/zer/DeepSeek-V2-Lite/ \
          --dataset-name "sonnet" \
          --dataset-path "../sonnet_4x.txt" \
          --sonnet-input-len 512 \
          --sonnet-output-len 6 \
          --num-prompts 10 \
          --port 8000 \
          --request-rate 2

and the server 8100 (Prefill, kv_producer) threw an error and GPU memory dropped, while the Decode side was fine. The related log:

INFO 03-05 19:14:32 logger.py:39] Received request cmpl-63569808618442019478e41d2204970b-0: prompt: "<｜begin▁of▁sentence｜>User: Pick as many lines as you can from these poem lines:\n\nFROM fairest creatures we desire increase,\nThat thereby beauty's rose might never die,\nBut as the riper should by time decease,\nHis tender heir might bear his memory:\nBut thou, contracted to thine own bright eyes,\nFeed'st thy light'st flame with self-substantial fuel,\nMaking a famine where abundance lies,\nThyself thy foe, to thy sweet self too cruel.\nThou that art now the world's fresh ornament\nAnd only herald to the gaudy spring,\nWithin thine own bud buriest thy content\nAnd, tender churl, makest waste in niggarding.\nPity the world, or else this glutton be,\nThese poor rude lines of thy deceased lover,\nWhen, in disgrace with fortune and men's eyes,\nMuch liker than your painted counterfeit:\nWhich to repair should be thy chief desire.\nYet, do thy worst, old Time: despite thy wrong,\nAs he takes from you, I engraft you new.\nThat then I scorn to change my state with kings.\nWhen I behold the violet past prime,\nBut thy eternal summer shall not fade\nHow can I then be elder than thou art?\nNor can thy shame give physic to my grief;\nO, carve not with thy hours my love's fair brow,\nWithout this, folly, age and cold decay:\nAnd night doth nightly make grief's strength seem stronger.\nMore than that tongue that more hath more express'd.\nBut that thou none lovest is most evident;\nThough thou repent, yet I have still the loss:\nOr any of these all, or all, or more,\nAll losses are restored and sorrows end.\nTo dry the rain on my storm-beaten face,\nWhich erst from heat did canopy the herd,\nFor thee and for myself no quiet find.\nHiding thy bravery in their rotten smoke?\nTo witness duty, not to show my wit:\nAnd every fair from fair sometime declines,\nThou of thyself thy sweet self dost deceive.\n\n\nAssistant:", params: SamplingParams(n=1, presence_penalty=0.0, frequency_penalty=0.0, repetition_penalty=1.0, temperature=0.0, top_p=1.0, top_k=-1, min_p=0.0, seed=None, stop=[], stop_token_ids=[], bad_words=[], include_stop_str_in_output=False, ignore_eos=False, max_tokens=1, min_tokens=0, logprobs=None, prompt_logprobs=None, skip_special_tokens=True, spaces_between_special_tokens=True, truncate_prompt_tokens=None, guided_decoding=None), prompt_token_ids: [100000, 100000, 5726, 25, 22982, 372, 1313, 4303, 372, 340, 481, 473, 1069, 20002, 4303, 25, 185, 185, 14047, 89394, 21774, 395, 8836, 4686, 11, 185, 4026, 19603, 8349, 6, 82, 8728, 1667, 1866, 6976, 11, 185, 2395, 372, 254, 4397, 524, 1023, 457, 766, 23483, 590, 11, 185, 11126, 15400, 35259, 1667, 8982, 586, 4887, 25, 185, 2395, 7852, 11, 38349, 276, 33714, 1377, 6464, 3545, 11, 185, 27872, 6, 292, 8060, 2156, 6, 292, 24138, 366, 1791, 12, 99051, 12692, 11, 185, 40011, 245, 61571, 1066, 20702, 11012, 11, 185, 1081, 42782, 8060, 39085, 11, 276, 8060, 6584, 1791, 1653, 17426, 13, 185, 37677, 344, 2460, 1132, 254, 1843, 6, 82, 5583, 31861, 185, 2280, 889, 53189, 276, 254, 36654, 88, 8079, 11, 185, 32130, 33714, 1377, 6581, 5446, 10399, 8060, 3093, 185, 2280, 11, 15400, 496, 3528, 11, 52922, 371, 9974, 279, 291, 28689, 18032, 13, 185, 47, 465, 254, 1843, 11, 410, 1979, 437, 1656, 35039, 330, 11, 185, 6723, 4922, 26663, 4303, 280, 8060, 40298, 20442, 11, 185, 3287, 11, 279, 36906, 366, 16531, 285, 1743, 6, 82, 3545, 11, 185, 30092, 2925, 250, 853, 520, 16786, 79129, 25, 185, 15526, 276, 9023, 1023, 330, 8060, 8037, 8836, 13, 185, 23014, 11, 536, 8060, 11312, 11, 1712, 6930, 25, 10988, 8060, 4360, 11, 185, 2124, 362, 4497, 473, 340, 11, 304, 2763, 3712, 340, 761, 13, 185, 4026, 937, 304, 31594, 276, 2262, 601, 1977, 366, 27198, 13, 185, 3287, 304, 26641, 254, 55067, 2882, 9966, 11, 185, 2395, 8060, 23553, 6237, 2786, 441, 28747, 185, 2819, 481, 304, 937, 330, 18418, 853, 7852, 2460, 30, 185, 27322, 481, 8060, 16635, 2102, 57335, 276, 601, 20110, 26, 185, 46, 11, 81103, 441, 366, 8060, 3821, 601, 2126, 6, 82, 4770, 28372, 11, 185, 21452, 437, 11, 36930, 11, 4500, 285, 6159, 11881, 25, 185, 2280, 2653, 33064, 59036, 1099, 20110, 6, 82, 6313, 1751, 13338, 13, 185, 7623, 853, 344, 16777, 344, 691, 15824, 691, 2856, 6, 67, 13, 185, 2395, 344, 7852, 5249, 645, 7034, 317, 1094, 11086, 26, 185, 23077, 7852, 37052, 11, 2618, 304, 463, 1592, 254, 4257, 25, 185, 3171, 688, 280, 1069, 521, 11, 410, 521, 11, 410, 691, 11, 185, 3315, 16384, 418, 19659, 285, 60248, 1225, 13, 185, 1898, 6989, 254, 9748, 331, 601, 10955, 12, 60199, 3059, 11, 185, 15526, 7455, 292, 473, 6142, 1216, 50905, 254, 38375, 11, 185, 1988, 9397, 285, 327, 4367, 642, 7257, 1275, 13, 185, 39, 3645, 8060, 63145, 279, 704, 58490, 14463, 30, 185, 1898, 10672, 9128, 11, 441, 276, 1296, 601, 20393, 25, 185, 2280, 1131, 4770, 473, 4770, 32422, 66878, 11, 185, 37677, 280, 60949, 8060, 6584, 1791, 46912, 51347, 13, 185, 185, 185, 77398, 25], lora_request: None, prompt_adapter_request: None.
INFO:     127.0.0.1:33560 - "POST /v1/completions HTTP/1.1" 200 OK
INFO 03-05 19:14:32 engine.py:275] Added request cmpl-63569808618442019478e41d2204970b-0.
ERROR 03-05 19:14:32 engine.py:139] AttributeError("'DeepseekV2Model' object has no attribute 'config'")
ERROR 03-05 19:14:32 engine.py:139] Traceback (most recent call last):
ERROR 03-05 19:14:32 engine.py:139]   File "/usr/local/lib/python3.10/dist-packages/vllm/engine/multiprocessing/engine.py", line 137, in start
ERROR 03-05 19:14:32 engine.py:139]     self.run_engine_loop()
ERROR 03-05 19:14:32 engine.py:139]   File "/usr/local/lib/python3.10/dist-packages/vllm/engine/multiprocessing/engine.py", line 200, in run_engine_loop
ERROR 03-05 19:14:32 engine.py:139]     request_outputs = self.engine_step()
ERROR 03-05 19:14:32 engine.py:139]   File "/usr/local/lib/python3.10/dist-packages/vllm/engine/multiprocessing/engine.py", line 218, in engine_step
ERROR 03-05 19:14:32 engine.py:139]     raise e
ERROR 03-05 19:14:32 engine.py:139]   File "/usr/local/lib/python3.10/dist-packages/vllm/engine/multiprocessing/engine.py", line 209, in engine_step
ERROR 03-05 19:14:32 engine.py:139]     return self.engine.step()
ERROR 03-05 19:14:32 engine.py:139]   File "/usr/local/lib/python3.10/dist-packages/vllm/engine/llm_engine.py", line 1386, in step
ERROR 03-05 19:14:32 engine.py:139]     outputs = self.model_executor.execute_model(
ERROR 03-05 19:14:32 engine.py:139]   File "/usr/local/lib/python3.10/dist-packages/vllm/executor/executor_base.py", line 138, in execute_model
ERROR 03-05 19:14:32 engine.py:139]     output = self.collective_rpc("execute_model",
ERROR 03-05 19:14:32 engine.py:139]   File "/usr/local/lib/python3.10/dist-packages/vllm/executor/uniproc_executor.py", line 51, in collective_rpc
ERROR 03-05 19:14:32 engine.py:139]     answer = run_method(self.driver_worker, method, args, kwargs)
ERROR 03-05 19:14:32 engine.py:139]   File "/usr/local/lib/python3.10/dist-packages/vllm/utils.py", line 2220, in run_method
ERROR 03-05 19:14:32 engine.py:139]     return func(*args, **kwargs)
ERROR 03-05 19:14:32 engine.py:139]   File "/usr/local/lib/python3.10/dist-packages/vllm/worker/worker_base.py", line 413, in execute_model
ERROR 03-05 19:14:32 engine.py:139]     output = self.model_runner.execute_model(
ERROR 03-05 19:14:32 engine.py:139]   File "/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py", line 116, in decorate_context
ERROR 03-05 19:14:32 engine.py:139]     return func(*args, **kwargs)
ERROR 03-05 19:14:32 engine.py:139]   File "/usr/local/lib/python3.10/dist-packages/vllm/worker/model_runner.py", line 1736, in execute_model
ERROR 03-05 19:14:32 engine.py:139]     get_kv_transfer_group().send_kv_caches_and_hidden_states(
ERROR 03-05 19:14:32 engine.py:139]   File "/usr/local/lib/python3.10/dist-packages/vllm/distributed/kv_transfer/kv_transfer_agent.py", line 61, in send_kv_caches_and_hidden_states
ERROR 03-05 19:14:32 engine.py:139]     self.connector.send_kv_caches_and_hidden_states(
ERROR 03-05 19:14:32 engine.py:139]   File "/usr/local/lib/python3.10/dist-packages/vllm/distributed/kv_transfer/kv_connector/simple_connector.py", line 165, in send_kv_caches_and_hidden_states
ERROR 03-05 19:14:32 engine.py:139]     model_config = model_executable.model.config
ERROR 03-05 19:14:32 engine.py:139]   File "/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py", line 1931, in __getattr__
ERROR 03-05 19:14:32 engine.py:139]     raise AttributeError(
ERROR 03-05 19:14:32 engine.py:139] AttributeError: 'DeepseekV2Model' object has no attribute 'config'
ERROR 03-05 19:14:43 client.py:300] RuntimeError('Engine process (pid 73163) died.')
ERROR 03-05 19:14:43 client.py:300] NoneType: None

When I located to line 165 of vllm/distributed/kv_transfer/kv_connector/simple_connector.py, the code was below:

model_config = model_executable.model.config
        num_heads = int(model_config.num_key_value_heads / self.tp_size)
        hidden_size = model_config.hidden_size
        num_attention_heads = model_config.num_attention_heads
        head_size = int(hidden_size / num_attention_heads)

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.