A decensored ultra-lightweight 0.6B Qwen3 model for resource-constrained environments. Made by RACER IS OP.

tools

ollama run R4C3R/qwen3-0.6b-heretic

curl http://localhost:11434/api/chat \
  -d '{
    "model": "R4C3R/qwen3-0.6b-heretic",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='R4C3R/qwen3-0.6b-heretic',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'R4C3R/qwen3-0.6b-heretic',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated yesterday

yesterday

11091b80cd22 · 397MB ·

model

archqwen3

parameters596M

quantizationQ4_K_M

397MB

template

{{- if .Suffix }}<|fim_prefix|>{{ .Prompt }}<|fim_suffix|>{{ .Suffix }}<|fim_middle|> {{- else if .M

1.6kB

Readme

🏴 Qwen3-0.6B-Heretic

Heretic Series by RACER IS OP

Qwen3-0.6B-Heretic is a decensored variant of Qwen’s ultra-small Qwen3 base. Features the same thinking/non-thinking dual-mode architecture as its larger siblings, without refusal guardrails. Tiny enough for any hardware.

Who this is for — developers who want Qwen3’s reasoning modes in the smallest possible package. Ideal for on-device reasoning agents, ultra-lightweight deployment, and resource-constrained environments.

Quickstart

ollama run R4C3R/qwen3-0.6b-heretic

VRAM Guide (choose your quant)

Quant	VRAM	When to use
`q4_k_m`	~0.4 GB	Best balance — runs on anything
`q8_0`	~0.6 GB	Maximum quality

This model fits anywhere. q8_0 is recommended since even the highest quality quant uses under 1 GB.

Links

sdad.pro · Heretic

License: Apache 2.0

Made with love by RACER IS OP — follow for more uncensored models in the Heretic Series.

# 🏴 Qwen3-0.6B-Heretic

> **Heretic Series** by RACER IS OP

---

Qwen3-0.6B-Heretic is a decensored variant of Qwen's ultra-small Qwen3 base. Features the same thinking/non-thinking dual-mode architecture as its larger siblings, without refusal guardrails. Tiny enough for any hardware.

**Who this is for** — developers who want Qwen3's reasoning modes in the smallest possible package. Ideal for on-device reasoning agents, ultra-lightweight deployment, and resource-constrained environments.

---

### Quickstart

```bash
ollama run R4C3R/qwen3-0.6b-heretic
```

### VRAM Guide (choose your quant)

| Quant | VRAM | When to use |
|-------|------|-------------|
| `q4_k_m` | ~0.4 GB | Best balance — runs on anything |
| `q8_0` | ~0.6 GB | Maximum quality |

> This model fits anywhere. `q8_0` is recommended since even the highest quality quant uses under 1 GB.

### Links

[sdad.pro](https://sdad.pro) · [Heretic](https://github.com/p-e-w/heretic)

**License:** [Apache 2.0](https://huggingface.co/Qwen/Qwen3-0.6B-Base/blob/main/LICENSE)

---

*Made with love by RACER IS OP — follow for more uncensored models in the Heretic Series.*

Paste, drop or click to upload images (.png, .jpeg, .jpg, .svg, .gif)