huihui_ai/huihui-4

huihui_ai/ huihui-4

644 Downloads Updated 1 week ago

Huihui4-8B-A4B is a lightweight MoE (Mixture of Experts) conversational model optimized from Google's gemma-4-26B-A4B-it architecture

vision tools thinking 8b

ollama run huihui_ai/huihui-4:8b

curl http://localhost:11434/api/chat \
  -d '{
    "model": "huihui_ai/huihui-4:8b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='huihui_ai/huihui-4:8b',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'huihui_ai/huihui-4:8b',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Applications

Claude Code ollama launch claude --model huihui_ai/huihui-4:8b

OpenClaw ollama launch openclaw --model huihui_ai/huihui-4:8b

Hermes Agent ollama launch hermes --model huihui_ai/huihui-4:8b

Codex ollama launch codex --model huihui_ai/huihui-4:8b

OpenCode ollama launch opencode --model huihui_ai/huihui-4:8b

Models

View all →

Name

4 models

Size

Context

Input

huihui-4:8b

6.6GB · 256K context window · Text, Image · 1 week ago

huihui-4:8b

6.6GB

256K

Text, Image

Readme

Huihui4-8B-A4B is a lightweight MoE (Mixture of Experts) conversational model optimized from Google’s gemma-4-26B-A4B-it architecture. Through expert pruning and supervised fine-tuning on high-quality dialogue data, this model significantly reduces computational overhead while preserving core reasoning and interaction capabilities. It is specifically designed for deployment on consumer-grade hardware and code-related conversational tasks.

This model is not an ablation variant.

Welcome everyone to test it.

References

[https://huggingface.co/huihui-ai/Huihui4-8B-A4B]