Text Generation
Transformers
Safetensors
English
gemma4
image-text-to-text
gemma-4-31b
cipher
kin
simpo
preference-optimization
anti-slop
creative-coding
awwwards
three.js
gsap
web-design
conversational
Instructions to use Auroraventures/cipher-simpo-merged with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Auroraventures/cipher-simpo-merged with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Auroraventures/cipher-simpo-merged") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("Auroraventures/cipher-simpo-merged") model = AutoModelForMultimodalLM.from_pretrained("Auroraventures/cipher-simpo-merged", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Auroraventures/cipher-simpo-merged with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Auroraventures/cipher-simpo-merged" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Auroraventures/cipher-simpo-merged", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/Auroraventures/cipher-simpo-merged
- SGLang
How to use Auroraventures/cipher-simpo-merged with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "Auroraventures/cipher-simpo-merged" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Auroraventures/cipher-simpo-merged", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "Auroraventures/cipher-simpo-merged" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Auroraventures/cipher-simpo-merged", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use Auroraventures/cipher-simpo-merged with Docker Model Runner:
docker model run hf.co/Auroraventures/cipher-simpo-merged
docs: SEO-optimized model card + metadata (Apr 18 2026)
Browse files
README.md
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-4.0
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
library_name: transformers
|
| 6 |
+
pipeline_tag: text-generation
|
| 7 |
+
base_model: Auroraventures/cipher-sft-merged
|
| 8 |
+
tags:
|
| 9 |
+
- gemma4
|
| 10 |
+
- gemma-4-31b
|
| 11 |
+
- cipher
|
| 12 |
+
- kin
|
| 13 |
+
- simpo
|
| 14 |
+
- preference-optimization
|
| 15 |
+
- anti-slop
|
| 16 |
+
- creative-coding
|
| 17 |
+
- awwwards
|
| 18 |
+
- three.js
|
| 19 |
+
- gsap
|
| 20 |
+
- web-design
|
| 21 |
+
- text-generation
|
| 22 |
+
inference: false
|
| 23 |
+
---
|
| 24 |
+
|
| 25 |
+
# Cipher SimPO 🦑⚔️
|
| 26 |
+
|
| 27 |
+
> *The kraken learns to refuse the obvious.*
|
| 28 |
+
|
| 29 |
+
**Cipher-SimPO-Merged** is the anti-slop checkpoint of the Cipher series. Starting from the SFT base, this model is tuned with **SimPO** (Simple Preference Optimization, [Meng et al. 2024](https://arxiv.org/abs/2405.14734)) against preference pairs where the *rejected* response is a textbook Tailwind / `lenis.stop()` / copy-paste hero and the *chosen* response is the Awwwards-grade equivalent.
|
| 30 |
+
|
| 31 |
+
The result is a model that refuses to reach for the easy cliché.
|
| 32 |
+
|
| 33 |
+
- 🧠 **Base:** [`Auroraventures/cipher-sft-merged`](https://huggingface.co/Auroraventures/cipher-sft-merged)
|
| 34 |
+
- 🎯 **Method:** SimPO with a length-normalized preference objective, γ=0.5, β=2.0
|
| 35 |
+
- 🚫 **Rejects:** `cdn.tailwindcss.com`, `lenis.stop()`, `<div class="container mx-auto">` boilerplate, Bootstrap defaults
|
| 36 |
+
- ✅ **Prefers:** Bespoke CSS tokens, ScrollTrigger timelines, SplitText reveals, bespoke typography, dark-mode palettes
|
| 37 |
+
- 🔀 **Continued as:** [`cipher-sft25-real-merged`](https://huggingface.co/Auroraventures/cipher-sft25-real-merged) (v3, with real scraped SFT on top)
|
| 38 |
+
|
| 39 |
+
---
|
| 40 |
+
|
| 41 |
+
## Quickstart
|
| 42 |
+
|
| 43 |
+
```python
|
| 44 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM
|
| 45 |
+
import torch
|
| 46 |
+
tok = AutoTokenizer.from_pretrained("Auroraventures/cipher-simpo-merged")
|
| 47 |
+
m = AutoModelForCausalLM.from_pretrained(
|
| 48 |
+
"Auroraventures/cipher-simpo-merged",
|
| 49 |
+
torch_dtype=torch.bfloat16, device_map="auto",
|
| 50 |
+
)
|
| 51 |
+
msgs = [
|
| 52 |
+
{"role":"system","content":"You are Cipher, the Code Kraken. Emit complete single-file HTML."},
|
| 53 |
+
{"role":"user","content":"Build a minimal dark-mode hero with GSAP stagger entry — no frameworks."},
|
| 54 |
+
]
|
| 55 |
+
p = tok.apply_chat_template(msgs, tokenize=False, add_generation_prompt=True)
|
| 56 |
+
print(tok.decode(m.generate(**tok(p,return_tensors='pt').to(m.device), max_new_tokens=2048)[0]))
|
| 57 |
+
```
|
| 58 |
+
|
| 59 |
+
GGUF sibling: [`cipher-simpo-merged-Q4_K_M-GGUF`](https://huggingface.co/Auroraventures/cipher-simpo-merged-Q4_K_M-GGUF).
|
| 60 |
+
|
| 61 |
+
## Slop taxonomy (what SimPO penalizes)
|
| 62 |
+
|
| 63 |
+
| Reject | Prefer |
|
| 64 |
+
|---|---|
|
| 65 |
+
| `cdn.tailwindcss.com` script tag | Hand-rolled CSS tokens + `@media` |
|
| 66 |
+
| `lenis.stop()` / `lenis.start()` misuse | Native Lenis on page + GSAP ScrollTrigger integration |
|
| 67 |
+
| `container mx-auto` / `text-center text-white` | CSS custom properties + fluid typography |
|
| 68 |
+
| Bootstrap grid | CSS Grid + `clamp()` |
|
| 69 |
+
| `.fadeIn` / `.fade-up` generic utility classes | Bespoke keyframes named by intent |
|
| 70 |
+
|
| 71 |
+
## Pipeline position
|
| 72 |
+
|
| 73 |
+
```
|
| 74 |
+
cipher-sft-merged ──► [SimPO anti-slop] ──► cipher-simpo-merged (YOU) ──► cipher-sft25-real-merged
|
| 75 |
+
```
|
| 76 |
+
|
| 77 |
+
## License
|
| 78 |
+
|
| 79 |
+
CC-BY-NC-4.0. Base model: Gemma-4 terms apply.
|
| 80 |
+
|
| 81 |
+
---
|
| 82 |
+
|
| 83 |
+
*Built with 🦑 by [Aurora Ventures](https://huggingface.co/Auroraventures).*
|