Can it support local deployment of Qwen3-Embedding?

[maxin9966]

1. Ollama's performance is too poor.
2. Can it support local deployment of Qwen3-Embedding?
3. Can it support local deployment of the Qwen3 LLM, which is compatible with OpenAI?

[maxin9966]

vllm + Qwen3-Embedding

Requires vllm>=0.8.5

import torch
import vllm
from vllm import LLM
model = LLM(model="Qwen/Qwen3-Embedding-0.6B", task="embed")

[suqi95716aa]

no any developer maintain issus?

[zc277584121]

@maxin9966 @suqi95716aa sorry for the delay,
for vllm embedding models, you can install vllm+ qwen3 in the server side, and use OpenAI API interface to request it in the deepsearcher client side.
for ollama , this PR unlimits the model map setting, allowing users to use any ollama embedding models