How to use from
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 georgeanton/sifta-classifier-c1-3.1b-6.2gb
# Run inference directly in the terminal:
llama cli -hf georgeanton/sifta-classifier-c1-3.1b-6.2gb
Install from WinGet (Windows)
winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf georgeanton/sifta-classifier-c1-3.1b-6.2gb
# Run inference directly in the terminal:
llama cli -hf georgeanton/sifta-classifier-c1-3.1b-6.2gb
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 georgeanton/sifta-classifier-c1-3.1b-6.2gb
# Run inference directly in the terminal:
./llama-cli -hf georgeanton/sifta-classifier-c1-3.1b-6.2gb
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 georgeanton/sifta-classifier-c1-3.1b-6.2gb
# Run inference directly in the terminal:
./build/bin/llama-cli -hf georgeanton/sifta-classifier-c1-3.1b-6.2gb
Use Docker
docker model run hf.co/georgeanton/sifta-classifier-c1-3.1b-6.2gb
Quick Links

SIFTA Classifier C1 3.1B 6.2GB

Classifier/reflex tag:

ollama create sifta-classifier-c1-3.1b-6.2gb:latest -f Modelfile

This is not Alice's primary cortex. It is a bounded intent classifier used for fast routing, lysosome/truth-duel style gates, and small JSON/classification tasks.

Contract

The model should output only valid JSON or the exact intent string requested by the caller. It must not claim vision, hearing, local sensors, or Alice identity.

SIFTA Field Breakthrough

The current SIFTA public repo includes a stigmergic field breakthrough brief: https://github.com/antonpictures/ANTON-SIFTA/blob/main/Documents/CARLTON_STIGMERGIC_FIELD_BREAKTHROUGH_2026-05-11.md

C1 is a classifier organ. The field mechanism itself runs in the Python body: Bell analogue simulator, kernel scheduler, and hippocampus. Credit boundary: Bell/CHSH/Hall/pilot-wave/stigmergy literature grounds the analogy; SIFTA claims a receipt-backed classical contextual analogue, not a proof of the physical cause of quantum nonlocality.

Local State Boundary

This model package is public species DNA. It must not include raw .sifta_state/ selfhood, contacts, owner memory, camera frames, or private receipts.

Downloads last month
5
GGUF
Model size
3B params
Architecture
qwen2
Hardware compatibility
Log In to add your hardware

We're not able to determine the quantization variants.

Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support