deepseek-ai

/

deepseek-vl2

like 384

Follow

DeepSeek 133k

Got the error below:
ValueError: The checkpoint you are trying to load has model type deepseek_vl_v2 but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date.

same error when deploying using vllm. Did you manage to solve? Thanks

Got the error below:
ValueError: The checkpoint you are trying to load has model type deepseek_vl_v2 but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date.

same error when deploying using vllm. Did you manage to solve? Thanks

Hi, Have you solved the problem, Thanks

Got the error below:
ValueError: The checkpoint you are trying to load has model type deepseek_vl_v2 but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date.

same error when deploying using vllm. Did you manage to solve? Thanks

Hi, Have you solved the problem, Thanks

same

I also got the same error

Transformers supported model list does not contain DeepSeek. You will have to import the model provided by DeepSeek source code

import torch
from transformers import AutoModelForCausalLM

from deepseek_vl.models import DeepseekVLV2Processor, DeepseekVLV2ForCausalLM

# specify the path to the model
model_path = "deepseek-ai/deepseek-vl2-small"
vl_chat_processor: DeepseekVLV2Processor = DeepseekVLV2Processor.from_pretrained(model_path)
tokenizer = vl_chat_processor.tokenizer

model = AutoModelForCausalLM.from_pretrained(model_path, trust_remote_code=True)

After upgrade vllm to the latest version e.g. v0.7.2, and run vllm with option --hf_overrides '{"architectures": ["DeepseekVLV2ForCausalLM"]}', then everything is OK.