ZimaBlueAI/HY-MT1.5-1.8:7b

ZimaBlueAI/ HY-MT1.5-1.8:7b

174 Downloads Updated 1 month ago

Hunyuan Translation Model Version 1.5 includes a 1.8B translation model. The 1.8B model can be deployed on edge devices and support real-time translation scenarios, making it widely applicable.

1.8b 7b

ollama run ZimaBlueAI/HY-MT1.5-1.8:7b

curl http://localhost:11434/api/chat \
  -d '{
    "model": "ZimaBlueAI/HY-MT1.5-1.8:7b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='ZimaBlueAI/HY-MT1.5-1.8:7b',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'ZimaBlueAI/HY-MT1.5-1.8:7b',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 1 month ago

1 month ago

a731e834aa97 · 4.6GB ·

model

archhunyuan-dense

parameters7.5B

quantizationQ4_K_M

4.6GB

template

{{ $has_head := true }} {{ if .System }}<|startoftext|>{{ .System }}<|extra_4|>{{ else }}{{ $has_hea

419B

license

TENCENT HY COMMUNITY LICENSE AGREEMENT Tencent HY-MT1.5 Release Date: December 30, 2025 THIS LICENSE

16kB

params

{ "repeat_penalty": 1 }

21B

Readme

This is an uncensored version of tencent/HY-MT1.5-1.8B created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.

If it’s only for translation, use the original model without ablation. If it involves translation and other conversations, the ablated model can be used.

ollama