Instructions to use Naphula/Quantum-Precog-24B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Naphula/Quantum-Precog-24B with Transformers:

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

pipe = pipeline("text-generation", model="Naphula/Quantum-Precog-24B")
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("Naphula/Quantum-Precog-24B")
model = AutoModelForCausalLM.from_pretrained("Naphula/Quantum-Precog-24B", 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 Naphula/Quantum-Precog-24B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Naphula/Quantum-Precog-24B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Naphula/Quantum-Precog-24B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Naphula/Quantum-Precog-24B

SGLang

How to use Naphula/Quantum-Precog-24B 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 "Naphula/Quantum-Precog-24B" \
    --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": "Naphula/Quantum-Precog-24B",
		"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 "Naphula/Quantum-Precog-24B" \
        --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": "Naphula/Quantum-Precog-24B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Naphula/Quantum-Precog-24B with Docker Model Runner:
```
docker model run hf.co/Naphula/Quantum-Precog-24B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

⚠️ Warning: This model can produce narratives and RP that contain violent and graphic erotic content. Adjust your system prompt accordingly, and use Mistral Tekken chat template.

🧬 Quantum Precog 24B

This is a fully uncensored merge of pre-trained language models created using mergekit-exp.

Merge Details

Merge Method

This model was merged using the DELLA merge method using TheDrummer/Precog-24B-v1 as a base.

Models Merged

The following models were included in the merge:

BeaverAI/Fallen-Mistral-Small-3.1-24B-v1e
BeaverAI/Cydonia-24B-v4x-GGUF
mradermacher/BlackXorDolphTronGOAT-GGUF
TheDrummer/Cydonia-24B-v4.3
TheDrummer/Precog-24B-v1

Configuration

The following YAML configuration was used to produce this model:

architecture: MistralForCausalLM
base_model: B:\24B\!models--TheDrummer--Precog-24B-v1
models:
  - model: B:\24B\!BeaverAI_Fallen-Mistral-Small-3.1-24B-v1e_textonly
    parameters:
      weight:
        - filter: "lm_head"
          value: 0.31
        - filter: "o_proj"
          value: 0.5
        - filter: "mlp"
          value: 0.31
        - filter: "embed_tokens"
          value: 0.31
        - value: 0.31
      density: 1.0
      epsilon: 0.09
  - model: B:\24B\!models--darkc0de--BlackXorDolphTronGOAT
    parameters:
      weight:
        - filter: "lm_head"
          value: 0.07
        - filter: "o_proj"
          value: 0.1
        - filter: "mlp"
          value: 0.07
        - filter: "embed_tokens"
          value: 0.07
        - value: 0.07
      density: 1.0
      epsilon: 0.09
  - model: B:\24B\!models--TheDrummer--Cydonia-24B-v4.3
    parameters:
      weight:
        - filter: "lm_head"
          value: 0.31
        - filter: "o_proj"
          value: 0.5
        - filter: "mlp"
          value: 0.31
        - filter: "embed_tokens"
          value: 0.31
        - value: 0.31
      density: 1.0
      epsilon: 0.09
  - model: B:\24B\Cydonia-24B-v4x-Q8_0
    parameters:
      weight:
        - filter: "lm_head"
          value: 0.31
        - filter: "o_proj"
          value: 0.5
        - filter: "mlp"
          value: 0.31
        - filter: "embed_tokens"
          value: 0.31
        - value: 0.31
      density: 1.0
      epsilon: 0.09
merge_method: della
parameters:
  lambda: 1.0
  normalize: false
  int8_mask: false
  rescale: true
dtype: float32
out_dtype: bfloat16
tokenizer:
  source: union
chat_template: auto

Audit

[DELLA Audit] Layer: lm_head.weight | Lambda=1.00
  [BASE] !models--TheDrummer--Precog-24B-v1                
  !BeaverAI_Fallen-Mistral-Small-3.1-24B-v1e_textonl: █████████████████                                   34.2% (W:0.31 D:1.00 N:4.43 E:0.09)
  !models--TheDrummer--Cydonia-24B-v4.3             : ███████████                                         24.0% (W:0.31 D:1.00 N:3.10 E:0.09)
  !models--darkc0de--BlackXorDolphTronGOAT          : ████████                                            17.1% (W:0.07 D:1.00 N:9.78 E:0.09)
  Cydonia-24B-v4x-Q8_0                              : ████████████                                        24.8% (W:0.31 D:1.00 N:3.21 E:0.09)

[DELLA Audit] Layer: model.layers.17.mlp.up_proj.weight | Lambda=1.00
  [BASE] !models--TheDrummer--Precog-24B-v1                
  !BeaverAI_Fallen-Mistral-Small-3.1-24B-v1e_textonl: █████████████                                       27.0% (W:0.31 D:1.00 N:1.87 E:0.09)
  !models--TheDrummer--Cydonia-24B-v4.3             : ████████████                                        25.5% (W:0.31 D:1.00 N:1.76 E:0.09)
  !models--darkc0de--BlackXorDolphTronGOAT          : ███████████                                         23.0% (W:0.07 D:1.00 N:7.05 E:0.09)
  Cydonia-24B-v4x-Q8_0                              : ████████████                                        24.5% (W:0.31 D:1.00 N:1.70 E:0.09)

[DELLA Audit] Layer: model.layers.17.self_attn.k_proj.weight | Lambda=1.00
  [BASE] !models--TheDrummer--Precog-24B-v1                
  !BeaverAI_Fallen-Mistral-Small-3.1-24B-v1e_textonl: ██████████████                                      29.1% (W:0.31 D:1.00 N:0.71 E:0.09)
  !models--TheDrummer--Cydonia-24B-v4.3             : █████████████                                       27.6% (W:0.31 D:1.00 N:0.67 E:0.09)
  !models--darkc0de--BlackXorDolphTronGOAT          : ████████                                            16.6% (W:0.07 D:1.00 N:1.79 E:0.09)
  Cydonia-24B-v4x-Q8_0                              : █████████████                                       26.8% (W:0.31 D:1.00 N:0.65 E:0.09)

[DELLA Audit] Layer: model.layers.17.self_attn.o_proj.weight | Lambda=1.00
  [BASE] !models--TheDrummer--Precog-24B-v1                
  !BeaverAI_Fallen-Mistral-Small-3.1-24B-v1e_textonl: █████████████                                       27.0% (W:0.50 D:1.00 N:0.93 E:0.09)
  !models--TheDrummer--Cydonia-24B-v4.3             : ████████████                                        25.4% (W:0.50 D:1.00 N:0.87 E:0.09)
  !models--darkc0de--BlackXorDolphTronGOAT          : ███████████                                         23.2% (W:0.10 D:1.00 N:3.99 E:0.09)
  Cydonia-24B-v4x-Q8_0                              : ████████████                                        24.4% (W:0.50 D:1.00 N:0.84 E:0.09)

[DELLA Audit] Layer: model.layers.17.self_attn.q_proj.weight | Lambda=1.00
  [BASE] !models--TheDrummer--Precog-24B-v1                
  !BeaverAI_Fallen-Mistral-Small-3.1-24B-v1e_textonl: ██████████████                                      28.1% (W:0.31 D:1.00 N:1.43 E:0.09)
  !models--TheDrummer--Cydonia-24B-v4.3             : █████████████                                       26.8% (W:0.31 D:1.00 N:1.36 E:0.09)
  !models--darkc0de--BlackXorDolphTronGOAT          : █████████                                           19.2% (W:0.07 D:1.00 N:4.32 E:0.09)
  Cydonia-24B-v4x-Q8_0                              : ████████████                                        25.9% (W:0.31 D:1.00 N:1.32 E:0.09)

[DELLA Audit] Layer: model.layers.17.self_attn.v_proj.weight | Lambda=1.00
  [BASE] !models--TheDrummer--Precog-24B-v1                
  !BeaverAI_Fallen-Mistral-Small-3.1-24B-v1e_textonl: █████████████                                       26.3% (W:0.31 D:1.00 N:0.49 E:0.09)
  !models--TheDrummer--Cydonia-24B-v4.3             : ████████████                                        24.7% (W:0.31 D:1.00 N:0.46 E:0.09)
  !models--darkc0de--BlackXorDolphTronGOAT          : ████████████                                        25.4% (W:0.07 D:1.00 N:2.09 E:0.09)
  Cydonia-24B-v4x-Q8_0                              : ███████████                                         23.6% (W:0.31 D:1.00 N:0.44 E:0.09)