mannix/omnimerge-v4:vision-IQ2

mannix/ omnimerge-v4:vision-IQ2_XXS

1,049 Downloads Updated 3 weeks ago

Qwen/Qwen3.6-27B + 3 Qwen3.6 fine-tunes with MLP-passthrough surgery

vision tools thinking

ollama run mannix/omnimerge-v4:vision-IQ2_XXS

curl http://localhost:11434/api/chat \
  -d '{
    "model": "mannix/omnimerge-v4:vision-IQ2_XXS",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='mannix/omnimerge-v4:vision-IQ2_XXS',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'mannix/omnimerge-v4:vision-IQ2_XXS',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 3 weeks ago

3 weeks ago

22285a02fc1f · 9.4GB

model

archqwen35

parameters26.9B

quantizationIQ2_XXS

8.4GB

projector

archclip

parameters461M

quantizationF16

928MB

params

{ "draft_num_predict": 3, "min_p": 0, "presence_penalty": 0, "repeat_penalty": 1,

114B

Readme

Model Card:

https://huggingface.co/ManniX-ITA/Qwen3.6-27B-Omnimerge-v4