vllm: [Usage]: Cannot register custom model (Out-of-Tree Model Integration)
Your current environment
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 22.04.5 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: version 3.22.1
Libc version: glibc-2.35
Python version: 3.10.12 (main, Nov 6 2024, 20:22:13) [GCC 11.4.0] (64-bit runtime)
Python platform: Linux-6.1.0-25-amd64-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 12.1.105
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration:
GPU 0: NVIDIA A100 80GB PCIe
GPU 1: NVIDIA A100 80GB PCIe
Nvidia driver version: 560.35.03
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.8.9.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.9.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.9.0
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.9.0
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.9.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.9.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.9.0
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
Address sizes: 46 bits physical, 57 bits virtual
Byte Order: Little Endian
CPU(s): 128
On-line CPU(s) list: 0-127
Vendor ID: GenuineIntel
Model name: Intel(R) Xeon(R) Gold 6338 CPU @ 2.00GHz
CPU family: 6
Model: 106
Thread(s) per core: 2
Core(s) per socket: 32
Socket(s): 2
Stepping: 6
CPU max MHz: 3200.0000
CPU min MHz: 800.0000
BogoMIPS: 4000.00
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 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 invpcid_single intel_ppin 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 split_lock_detect wbnoinvd dtherm ida arat pln pts avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme avx512_vpopcntdq la57 rdpid fsrm md_clear pconfig flush_l1d arch_capabilities
Virtualization: VT-x
L1d cache: 3 MiB (64 instances)
L1i cache: 2 MiB (64 instances)
L2 cache: 80 MiB (64 instances)
L3 cache: 96 MiB (2 instances)
NUMA node(s): 2
NUMA node0 CPU(s): 0-31,64-95
NUMA node1 CPU(s): 32-63,96-127
Vulnerability Gather data sampling: Mitigation; Microcode
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Mitigation; Clear CPU buffers; SMT vulnerable
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; RSB filling; PBRSB-eIBRS SW sequence; BHI SW loop, KVM SW loop
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Not affected
Versions of relevant libraries:
[pip3] numpy==2.2.0
[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-ml-py==12.560.30
[pip3] nvidia-nccl-cu12==2.21.5
[pip3] nvidia-nvjitlink-cu12==12.4.127
[pip3] nvidia-nvtx-cu12==12.4.127
[pip3] pynvml==11.5.3
[pip3] pyzmq==26.2.0
[pip3] torch==2.5.1
[pip3] torchvision==0.20.1
[pip3] transformers==4.47.0
[pip3] triton==3.1.0
[conda] Could not collect
ROCM Version: Could not collect
Neuron SDK Version: N/A
vLLM Version: 0.6.4.post1
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0 GPU1 NIC0 NIC1 CPU Affinity NUMA Affinity GPU NUMA ID
GPU0 X PIX NODE NODE 32-63,96-127 1 N/A
GPU1 PIX X NODE NODE 32-63,96-127 1 N/A
NIC0 NODE NODE X PIX
NIC1 NODE NODE 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
NVIDIA_VISIBLE_DEVICES=/var/run/nvidia-container-devices
NVIDIA_REQUIRE_CUDA=cuda>=12.1 brand=tesla,driver>=470,driver<471 brand=unknown,driver>=470,driver<471 brand=nvidia,driver>=470,driver<471 brand=nvidiartx,driver>=470,driver<471 brand=geforce,driver>=470,driver<471 brand=geforcertx,driver>=470,driver<471 brand=quadro,driver>=470,driver<471 brand=quadrortx,driver>=470,driver<471 brand=titan,driver>=470,driver<471 brand=titanrtx,driver>=470,driver<471 brand=tesla,driver>=525,driver<526 brand=unknown,driver>=525,driver<526 brand=nvidia,driver>=525,driver<526 brand=nvidiartx,driver>=525,driver<526 brand=geforce,driver>=525,driver<526 brand=geforcertx,driver>=525,driver<526 brand=quadro,driver>=525,driver<526 brand=quadrortx,driver>=525,driver<526 brand=titan,driver>=525,driver<526 brand=titanrtx,driver>=525,driver<526
NCCL_VERSION=2.17.1-1
NVIDIA_DRIVER_CAPABILITIES=compute,utility
NVIDIA_PRODUCT_NAME=CUDA
NVIDIA_CUDA_END_OF_LIFE=1
CUDA_VERSION=12.1.0
LD_LIBRARY_PATH=/home/pokonechnyy.e/.local/lib/python3.10/site-packages/cv2/../../lib64:
CUDA_MODULE_LOADING=LAZY
How would you like to use vllm
I want to run slightly modified version of Qwen2.5-32B-Instruct (to be more precise, I just want to add bias term to lm_head, original Qwen has only lm_head.weight and no bias). I want to avoid forking vLLM and tried to follow documentation (point 6 “Out-of-Tree Model Integration”)
To start off simple, I decided to just inherit the class for the original Qwen2 and try to register it as a custom model. So I wrapped a file named like qwen2biased.py with content:
# qwen2biased.py
import torch
import torch.nn as nn
from typing import Optional, List, Union, Tuple
from transformers import Qwen2Config, Qwen2Model, Qwen2Tokenizer, Qwen2PreTrainedModel, GenerationMixin
from transformers.modeling_outputs import CausalLMOutputWithPast
from vllm.model_executor.models.qwen2 import Qwen2ForCausalLM
from vllm.config import CacheConfig, VllmConfig
class Qwen2ForCausalLMBiased(Qwen2ForCausalLM):
#
# NO CHANGES in init, just straight copy from
# https://github.com/vllm-project/vllm/blob/b3b1526f03906c935e6ef80a2cdc971a65fdf7e2/vllm/model_executor/models/qwen2.py#L401
def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
super().__init__()
config = vllm_config.model_config.hf_config
quant_config = vllm_config.quant_config
lora_config = vllm_config.lora_config
self.config = config
self.lora_config = lora_config
self.quant_config = quant_config
self.model = Qwen2Model(vllm_config=vllm_config,
prefix=maybe_prefix(prefix, "model"))
if get_pp_group().is_last_rank:
if config.tie_word_embeddings:
self.lm_head = self.model.embed_tokens
else:
self.lm_head = ParallelLMHead(config.vocab_size,
config.hidden_size,
quant_config=quant_config,
prefix=maybe_prefix(
prefix, "lm_head"))
else:
self.lm_head = PPMissingLayer()
self.logits_processor = LogitsProcessor(config.vocab_size)
self.sampler = get_sampler()
self.make_empty_intermediate_tensors = (
self.model.make_empty_intermediate_tensors)
And then I want to replace entrypoint (as suggested in the docs) like this:
# custom_entrypoint.py
from vllm import ModelRegistry
if __name__ == '__main__':
from qwen2biased import Qwen2ForCausalLMBiased
ModelRegistry.register_model("Qwen2ForCausalLMBiased", Qwen2ForCausalLMBiased)
import runpy
runpy.run_module('vllm.entrypoints.openai.api_server', run_name='__main__')
And try to run like this:
python3 -O -u -m custom_entrypoint.py \
--host=127.0.0.1 \
--port=8000 \
--model=/path-to/Qwen2.5-32B-Instruct-NC-INT8 \
--served-model-name=qwen2-5-32b-instruct
where config.json with my model is
{
"_name_or_path": "Qwen2.5-32B-Instruct-NC-INT8",
"architectures": [
"Qwen2ForCausalLMBiased"
],
"model_type": "qwen2",
... # other unchanged
}
However, I’m still getting error:
Model architectures ['Qwen2ForCausalLMBiased'] are not supported for now. Supported architectures:
dict_keys(['AquilaModel', 'AquilaForCausalLM', 'ArcticForCausalLM', 'BaiChuanForCausalLM', 'BaichuanForCausalLM',
'BloomForCausalLM', 'CohereForCausalLM', 'DbrxForCausalLM', 'DeciLMForCausalLM', 'DeepseekForCausalLM',
'DeepseekV2ForCausalLM', 'ExaoneForCausalLM', 'FalconForCausalLM', 'GemmaForCausalLM', 'Gemma2ForCausalLM',
'GPT2LMHeadModel', 'GPTBigCodeForCausalLM', 'GPTJForCausalLM', 'GPTNeoXForCausalLM', 'GraniteForCausalLM',
'GraniteMoeForCausalLM', 'InternLMForCausalLM', 'InternLM2ForCausalLM', 'InternLM2VEForCausalLM',
'JAISLMHeadModel', 'JambaForCausalLM', 'LlamaForCausalLM', 'LLaMAForCausalLM', 'MambaForCausalLM',
'FalconMambaForCausalLM', 'MiniCPMForCausalLM', 'MiniCPM3ForCausalLM', 'MistralForCausalLM', 'MixtralForCausalLM',
'QuantMixtralForCausalLM', 'MptForCausalLM', 'MPTForCausalLM', 'NemotronForCausalLM', 'OlmoForCausalLM',
'OlmoeForCausalLM', 'OPTForCausalLM', 'OrionForCausalLM', 'PersimmonForCausalLM', 'PhiForCausalLM',
'Phi3ForCausalLM', 'Phi3SmallForCausalLM', 'PhiMoEForCausalLM', 'Qwen2ForCausalLM', 'Qwen2MoeForCausalLM',
'RWForCausalLM', 'StableLMEpochForCausalLM', 'StableLmForCausalLM', 'Starcoder2ForCausalLM', 'SolarForCausalLM',
'XverseForCausalLM', 'BartModel', 'BartForConditionalGeneration', 'Florence2ForConditionalGeneration', 'BertModel',
'RobertaModel', 'XLMRobertaModel', 'Gemma2Model', 'LlamaModel', 'MistralModel', 'Qwen2Model',
'Qwen2ForRewardModel', 'Qwen2ForSequenceClassification', 'LlavaNextForConditionalGeneration', 'Phi3VForCausalLM',
'Qwen2VLForConditionalGeneration', 'Blip2ForConditionalGeneration', 'ChameleonForConditionalGeneration',
'ChatGLMModel', 'ChatGLMForConditionalGeneration', 'FuyuForCausalLM', 'H2OVLChatModel', 'InternVLChatModel',
'Idefics3ForConditionalGeneration', 'LlavaForConditionalGeneration', 'LlavaNextVideoForConditionalGeneration',
'LlavaOnevisionForConditionalGeneration', 'MiniCPMV', 'MolmoForCausalLM', 'NVLM_D',
'PaliGemmaForConditionalGeneration', 'PixtralForConditionalGeneration', 'QWenLMHeadModel',
'Qwen2AudioForConditionalGeneration', 'UltravoxModel', 'MllamaForConditionalGeneration', 'EAGLEModel',
'MedusaModel', 'MLPSpeculatorPreTrainedModel'])
ERROR 12-16 09:51:31 engine.py:366] Traceback (most recent call last):
ERROR 12-16 09:51:31 engine.py:366] File "/home/pokonechnyy.e/.local/lib/python3.10/site-packages/vllm/engine/multiprocessing/engine.py", line 357, in run_mp_engine
ERROR 12-16 09:51:31 engine.py:366] engine = MQLLMEngine.from_engine_args(engine_args=engine_args,
ERROR 12-16 09:51:31 engine.py:366] File "/home/pokonechnyy.e/.local/lib/python3.10/site-packages/vllm/engine/multiprocessing/engine.py", line 114, in from_engine_args
ERROR 12-16 09:51:31 engine.py:366] engine_config = engine_args.create_engine_config()
ERROR 12-16 09:51:31 engine.py:366] File "/home/pokonechnyy.e/.local/lib/python3.10/site-packages/vllm/engine/arg_utils.py", line 959, in create_engine_config
ERROR 12-16 09:51:31 engine.py:366] model_config = self.create_model_config()
ERROR 12-16 09:51:31 engine.py:366] File "/home/pokonechnyy.e/.local/lib/python3.10/site-packages/vllm/engine/arg_utils.py", line 891, in create_model_config
ERROR 12-16 09:51:31 engine.py:366] return ModelConfig(
ERROR 12-16 09:51:31 engine.py:366] File "/home/pokonechnyy.e/.local/lib/python3.10/site-packages/vllm/config.py", line 251, in __init__
ERROR 12-16 09:51:31 engine.py:366] self.multimodal_config = self._init_multimodal_config(
ERROR 12-16 09:51:31 engine.py:366] File "/home/pokonechnyy.e/.local/lib/python3.10/site-packages/vllm/config.py", line 277, in _init_multimodal_config
ERROR 12-16 09:51:31 engine.py:366] if ModelRegistry.is_multimodal_model(architectures):
ERROR 12-16 09:51:31 engine.py:366] File "/home/pokonechnyy.e/.local/lib/python3.10/site-packages/vllm/model_executor/models/registry.py", line 422, in is_multimodal_model
ERROR 12-16 09:51:31 engine.py:366] return self.inspect_model_cls(architectures).supports_multimodal
ERROR 12-16 09:51:31 engine.py:366] File "/home/pokonechnyy.e/.local/lib/python3.10/site-packages/vllm/model_executor/models/registry.py", line 391, in inspect_model_cls
ERROR 12-16 09:51:31 engine.py:366] return self._raise_for_unsupported(architectures)
ERROR 12-16 09:51:31 engine.py:366] File "/home/pokonechnyy.e/.local/lib/python3.10/site-packages/vllm/model_executor/models/registry.py", line 352, in _raise_for_unsupported
ERROR 12-16 09:51:31 engine.py:366] raise ValueError(
(a lot of stuff but my model was not registered)
What I am doing wrong? Any help would be really appreciated!
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.
About this issue
- Original URL
- State: closed
- Created 2 years ago
- Comments: 27 (16 by maintainers)
yeah the code is outdated. please check https://docs.vllm.ai/en/latest/models/adding_model.html#out-of-tree-model-integration for the up-to-date doc, and also https://docs.vllm.ai/en/latest/design/plugin_system.html#what-can-plugins-do
Oh, I think this code is outdated. cc @youkaichao
Now, you should use plugins to register models when running vLLM in a separate process. See https://github.com/vllm-project/vllm/tree/main/tests/plugins/vllm_add_dummy_model for an example package.
You should change this to point to the absolute import path (so in your case, it would be
vllm_phi_custom.vllm_register_new_model:register). To simplify things, I suggest you move the function to__init__.pyright insidevllm_phi_custom/so you can change the import tovllm_phi_custom:registerdirectly.