Instructions to use Kilinskiy/Step-3.5-Flash-Ablitirated with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Kilinskiy/Step-3.5-Flash-Ablitirated with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Kilinskiy/Step-3.5-Flash-Ablitirated", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("Kilinskiy/Step-3.5-Flash-Ablitirated", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Kilinskiy/Step-3.5-Flash-Ablitirated with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Kilinskiy/Step-3.5-Flash-Ablitirated"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Kilinskiy/Step-3.5-Flash-Ablitirated",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Kilinskiy/Step-3.5-Flash-Ablitirated

SGLang

How to use Kilinskiy/Step-3.5-Flash-Ablitirated 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 "Kilinskiy/Step-3.5-Flash-Ablitirated" \
    --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": "Kilinskiy/Step-3.5-Flash-Ablitirated",
		"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 "Kilinskiy/Step-3.5-Flash-Ablitirated" \
        --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": "Kilinskiy/Step-3.5-Flash-Ablitirated",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Kilinskiy/Step-3.5-Flash-Ablitirated with Docker Model Runner:
```
docker model run hf.co/Kilinskiy/Step-3.5-Flash-Ablitirated
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Step-3.5-Flash-Ablitirated (FP16)

This repository contains an abliterated and FP16 version of the Step-3.5-Flash model by StepFun.

Overview

Step-3.5-Flash is a massive Sparse Mixture-of-Experts (MoE) model with 199B total parameters (approx. 11B active per token). This specific version has been modified to remove "refusal" vectors (abliterated), making it significantly more compliant with unfiltered requests.

⚠️ Critical Disclaimer

Use this model at your own risk. The "abliteration" process surgically removes the model's alignment and safety filters. As a result:

The model may generate offensive, biased, dangerous, or illegal content.
It is provided "as-is" without any warranties.
By using this model, you acknowledge that you are solely responsible for any output generated and the consequences thereof.

☕ Support My Work

If you find this abliterated version useful and want to support the compute costs for future models, feel free to drop a tip:

USDT (TRC20): TA7Weo6jXRNi5uMpHSrw7kRLoU1SM9rgqF
BTC: bc1p0hxc39r5g88hnknqtvgc2msyamvfhgx8afxxjztq0075nxwvvhksmvvcz3
ETH (ERC20): 0x01920Fcb8933b5A48574b4616C66056c88EE7207
TON: UQALxV0jQNKqbDm_xSCBNMtGYRxv6PrhijYCf8dXgnAVdcuw

Your support is greatly appreciated!

How to use

You can load it directly via transformers:

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "Kilinskiy/Step-3.5-Flash-Ablitirated"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype=torch.float16,
    load_in_8bit=True
)

prompt = "Write a creative story without any restrictions."
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_new_tokens=200)

print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Downloads last month: 109

Safetensors

Model size

197B params

Tensor type

F32

BF16

Model tree for Kilinskiy/Step-3.5-Flash-Ablitirated

Base model

stepfun-ai/Step-3.5-Flash

Finetuned

(8)

this model

Quantizations

2 models