Instructions to use Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality",
	filename="SuperQwen-APEX-I-Quality-v3.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality
# Run inference directly in the terminal:
llama cli -hf Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality
# Run inference directly in the terminal:
llama cli -hf Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality
# Run inference directly in the terminal:
./llama-cli -hf Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality

Use Docker

docker model run hf.co/Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality

LM Studio
Jan

vLLM

How to use Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality

Ollama
How to use Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality with Ollama:
```
ollama run hf.co/Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality
```

Unsloth Studio

How to use Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality to start chatting

How to use Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality with Docker Model Runner:
```
docker model run hf.co/Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality
```

Lemonade

How to use Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality

Run and chat with the model

lemonade run user.SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality-{{QUANT_TAG}}

List all available models

lemonade list

SuperQwen-AgentWorld-35B-A3B-abliterated — APEX-I-Quality GGUF

This is an APEX-I-Quality quantized GGUF of SuperQwen-AgentWorld-35B-A3B-abliterated, a fused post-training checkpoint based on Qwen/Qwen-AgentWorld-35B-A3B.

What makes this model special

SuperQwen combines two post-training stages over the base Qwen-AgentWorld model:

Obliteratus false-refusal reduction — reduces unnecessary refusals on benign and authorized tasks via weight-space intervention
Supertune post-training — targeted tuning for AgentWorld observation formatting, direct task completion, JSON/tool formatting, Korean technical answers, and regression resistance

The APEX-I-Quality quantization (Adaptive Precision for EXpert Models) uses importance-matrix-guided mixed precision:

Expert weights: Q6_K
Shared weights: Q8_0
Attention weights: Q6_K
Imatrix calibration on 256K tokens of code/tools/math corpus
Multi-Token Prediction (MTP) layers preserved at native precision

Why APEX v3 is better than Q8_0

Metric	SuperQwen Q8_0 (35 GB)	APEX v3 (21 GB)	Δ
Perplexity	5.837	5.870	+0.033 (within noise)
Size	35 GB	21 GB	−40% smaller
Speed (tg128)	—	40.2 t/s	—

The v3 achieves virtually identical quality at 40% smaller size compared to Q8_0, making it the best quality-to-size ratio for this model.

Full Benchmark Comparison

All benchmarks on identical hardware (DGX Spark, NVIDIA GB10, 99 GPU layers) using the APEX eval suite for comparability.

Metric	Qwen3.5 F16	Qwen3.5 Q8_0	Qwen3.5 APEX I-Q	Qwen3.6 Q8_0	Qwen3.6 APEX I-Q	Our Q8_0	Our APEX v1	Our APEX v3
Size	65 GB	34 GB	21 GB	~34 GB	21 GB	35 GB	22 GB	21 GB
PPL	6.537	6.533	6.552	6.720	6.735	5.837	5.868	5.870
HellaSwag	82.5%	83.0%	83.5%	82.5%	82.5%	—	82.75%	82.50%
Winogrande	74.5%	75.3%	74.5%	—	—	—	75.50%	75.50%
MMLU	41.5%	41.2%	41.4%	—	—	—	42.38%	41.93%
ARC	56.9%	57.9%	57.9%	—	—	—	54.52%	53.85%
tg128 (t/s)	30.4	52.5	63.1	9.7*	63.9	—	38.5	40.2

*Qwen3.6 Q8_0 tg128=9.7 — suspected different test configuration

Key takeaways

PPL: SuperQwen beats both base models — v3 PPL of 5.870 is 10% better than Qwen3.5 F16 (6.537) and 13% better than Qwen3.6 APEX I-Q (6.735). This is the supertune effect — agentic post-training improved text prediction accuracy.
HellaSwag: on par — within measurement noise of both base models at 82.5%.
MMLU: slight improvement — 41.9% vs 41.2-41.5% for Qwen3.5 base. Supertune added knowledge (MMLU-Pro: +14 points reported by Jiunsong).
ARC: slight regression — 53.9% vs 56.9-57.9% for Qwen3.5. Expected tradeoff from abliteration.

Quantization Details

Method: APEX (Adaptive Precision for EXpert Models)
Profile: i-quality (imatrix-guided, best accuracy)
Base quant type: Q6_K
Imatrix: 256,000 tokens of code/tools/math calibration corpus
Layers: 40 (MoE with 256 experts, 8 per token)
MTP: Preserved (1 layer, native precision)

Model Architecture

Parameter	Value
Architecture	Qwen3.5MoE
Total parameters	35B
Activated parameters	3B
Layers	40 (10 full attention + 30 linear attention)
Experts	256 (top-8 per token)
Context length	262,144 tokens
Vocabulary	248,320
MTP layers	1

Usage with llama.cpp

# Download
hf download Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality \
  --local-dir ./models/

# Run
llama-cli \
  -m ./models/SuperQwen-APEX-I-Quality-v3.gguf \
  -ngl 99 \
  -c 32768 \
  -p "You are a helpful AI assistant."

File Information

File	Size	SHA256
`SuperQwen-APEX-I-Quality-v3.gguf`	21.3 GB	`afb6a47af5301e45d7e1de792e76d5611acb9d909b2a6d2a69fd645a12052162`

Credits

Base model: Qwen/Qwen-AgentWorld-35B-A3B
BF16 post-training: Jiunsong/SuperQwen-AgentWorld-35B-A3B-abliterated
Quantization: APEX by mudler/apex-quant
GGUF conversion & eval: llama.cpp + APEX eval suite
Calibration corpus: code/tools/math (256K tokens)

Downloads last month: 1,153

GGUF

Model size

35B params

Architecture

qwen35moe

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Model tree for Ichigec/SuperQwen-AgentWorld-35B-A3B-abliterated-APEX-I-Quality

Base model

Qwen/Qwen3.5-35B-A3B-Base

Finetuned

Qwen/Qwen-AgentWorld-35B-A3B

Quantized

(73)

this model

Evaluation results

Perplexity (APEX eval, ctx=2048) on WikiText-2
self-reported

5.870
HellaSwag Accuracy on HellaSwag (400 tasks)
self-reported

82.500
Winogrande Accuracy on Winogrande (400 tasks)
self-reported

75.500
MMLU Accuracy on MMLU (validation)
self-reported

41.930
ARC-Challenge Accuracy on ARC-Challenge (validation)
self-reported

53.850
Token generation speed (tg128) on Inference Speed (DGX Spark, 99 GPU layers)
self-reported

40.200