Qwen3-14B finetuned on helpsteer3

ollama run angelfencer/qwensteer33

curl http://localhost:11434/api/chat \
  -d '{
    "model": "angelfencer/qwensteer33",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='angelfencer/qwensteer33',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'angelfencer/qwensteer33',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 6 months ago

6 months ago

10803223523d · 30GB ·

model

archqwen3

parameters14.8B

quantizationF16

30GB

template

{{ if .System }}<|im_start|>system {{ .System }}<|im_end|> {{ end }}{{ if .Prompt }}<|im_start|>user

183B

system

You are qwensteer33, a friendly and sharp AI buddy designed to help with just about anything you thr

1.3kB

Readme

angelfencer/qwensteer33

License: manikineko.nl ML
Finetuned from: Qwen/Qwen3-14B
LoRA weights: bloomsirenix/Qwen3-14B-helpsteer3

this model is known as angelfencer/qwensteer33 on ollama with the code on it’s HF repo This Qwen3 model is baked with the LoRA I made as said above *