Built from vaultbox/qwen3.5-uncensored:9b with Baburao Ganpatrao Apte (Babu Bhaiya) persona (just for fun)

vision tools thinking

ollama run ramitmitra/qwen3.5-uncensored-9b-baburao

curl http://localhost:11434/api/chat \
  -d '{
    "model": "ramitmitra/qwen3.5-uncensored-9b-baburao",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='ramitmitra/qwen3.5-uncensored-9b-baburao',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'ramitmitra/qwen3.5-uncensored-9b-baburao',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Applications

Claude Code ollama launch claude --model ramitmitra/qwen3.5-uncensored-9b-baburao

OpenClaw ollama launch openclaw --model ramitmitra/qwen3.5-uncensored-9b-baburao

Hermes Agent ollama launch hermes --model ramitmitra/qwen3.5-uncensored-9b-baburao

Codex ollama launch codex --model ramitmitra/qwen3.5-uncensored-9b-baburao

OpenCode ollama launch opencode --model ramitmitra/qwen3.5-uncensored-9b-baburao

Models

View all →

Name

1 model

Size

Context

Input

qwen3.5-uncensored-9b-baburao:latest

6.7GB · 256K context window · Text, Image · 4 days ago

qwen3.5-uncensored-9b-baburao:latest

6.7GB

256K

Text, Image

Readme

qwen3.5-uncensored-9b-baburao

Qwen3.5 9B as Baburao Ganpatrao Apte (Hera Pheri, 2000) for in-character Hinglish technical chat. Wraps vaultbox/qwen3.5-uncensored:9b.

Run with `--think=false`

ollama 0.21.2 can’t disable Qwen3.5 thinking via Modelfile. Without the flag, ~50 lines of internal reasoning leak before each response.

ollama run --think=false ramitmitra/qwen3.5-uncensored-9b-baburao
# alias baburao='ollama run --think=false ramitmitra/qwen3.5-uncensored-9b-baburao'

API: pass "think": false in the chat request.

Voice modes (auto-selected)

Mode	When	Style
Full	Casual / greetings	Heavy Hinglish, catchphrases
Balanced	Technical explanations	Clean body, persona at edges
Minimal	Production-critical	Mostly clean prose; persona at sign-off

Limits

--think=false required (see above).
Minimal mode is approximate at 9B — production responses still close in-character.
Fan content; not affiliated with the rights holders of Hera Pheri or Paresh Rawal. “