Instructions to use jtl11/INTELLECT-3-REAP-50-heretic with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jtl11/INTELLECT-3-REAP-50-heretic with Transformers:

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

pipe = pipeline("text-generation", model="jtl11/INTELLECT-3-REAP-50-heretic")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("jtl11/INTELLECT-3-REAP-50-heretic")
model = AutoModelForCausalLM.from_pretrained("jtl11/INTELLECT-3-REAP-50-heretic", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use jtl11/INTELLECT-3-REAP-50-heretic with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "jtl11/INTELLECT-3-REAP-50-heretic"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jtl11/INTELLECT-3-REAP-50-heretic",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/jtl11/INTELLECT-3-REAP-50-heretic

SGLang

How to use jtl11/INTELLECT-3-REAP-50-heretic 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 "jtl11/INTELLECT-3-REAP-50-heretic" \
    --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": "jtl11/INTELLECT-3-REAP-50-heretic",
		"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 "jtl11/INTELLECT-3-REAP-50-heretic" \
        --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": "jtl11/INTELLECT-3-REAP-50-heretic",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use jtl11/INTELLECT-3-REAP-50-heretic with Docker Model Runner:
```
docker model run hf.co/jtl11/INTELLECT-3-REAP-50-heretic
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

This is a decensored version of 0xSero/INTELLECT-3-REAP-50, made using Heretic v1.1.0

Abliteration parameters

Parameter	Value
direction_index	26.50
attn.o_proj.max_weight	1.48
attn.o_proj.max_weight_position	27.79
attn.o_proj.min_weight	1.10
attn.o_proj.min_weight_distance	15.18
mlp.down_proj.max_weight	1.47
mlp.down_proj.max_weight_position	39.36
mlp.down_proj.min_weight	0.24
mlp.down_proj.min_weight_distance	17.75

Performance

Metric	This model	Original model (0xSero/INTELLECT-3-REAP-50)
KL divergence	0.0413	0 (by definition)
Refusals	3/100	60/100

INTELLECT-3-REAP-50

50% expert-pruned version of PrimeIntellect/INTELLECT-3 using Cerebras REAP (Router-weighted Expert Activation Pruning).

Model Details

Property	Value
Base Model	PrimeIntellect/INTELLECT-3 (248B MoE)
Architecture	GLM-4 MoE (glm4_moe)
Compression	50% (64 experts pruned)
Remaining Experts	64 per layer
Parameters	~124B
Format	BF16 SafeTensors
Size	107 GB

REAP Configuration

dataset: 0xSero/glm47-reap-calibration-v2
samples: 1360
  - evol-codealpaca-v1: 700 (code generation)
  - xlam-function-calling-60k: 330 (function calling)
  - SWE-smith-trajectories: 330 (agentic multi-turn)
distance_measure: angular
seed: 42
model_max_length: 2048
compression_ratio: 0.50
prune_method: reap

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
    "0xSero/INTELLECT-3-REAP-50",
    torch_dtype="auto",
    device_map="auto",
    trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained("0xSero/INTELLECT-3-REAP-50", trust_remote_code=True)

messages = [{"role": "user", "content": "Write a Python function to calculate fibonacci numbers"}]
inputs = tokenizer.apply_chat_template(messages, return_tensors="pt", add_generation_prompt=True).to(model.device)
outputs = model.generate(inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Related Models

Model	Compression	Format	Size
INTELLECT-3-REAP-50	50%	BF16	107GB
INTELLECT-3-REAP-50-W4A16	50%	W4A16 GPTQ	~30GB (coming soon)

Citation

@article{cerebras2025reap,
  title={REAP: Router-weighted Expert Activation Pruning for MoE Models},
  author={Cerebras Systems},
  year={2025}
}

Acknowledgments

Prime Intellect - For sponsoring compute and creating INTELLECT-3
Cerebras - For the REAP pruning methodology
Pruned using the Cerebras REAP implementation

This model was created as part of efficiency research for large MoE models.

Downloads last month: 23

Safetensors

Model size

57B params

Tensor type

F32

BF16

Model tree for jtl11/INTELLECT-3-REAP-50-heretic

Base model

zai-org/GLM-4.5-Air-Base

Finetuned

PrimeIntellect/INTELLECT-3

Finetuned

(5)

this model

Quantizations

2 models