Instructions to use brsx-labs/BRSX-Reasoner with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- llama-cpp-python
How to use brsx-labs/BRSX-Reasoner with llama-cpp-python:
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="brsx-labs/BRSX-Reasoner", filename="BRSX-Reasoner.gguf", )
llm.create_chat_completion( messages = "No input example has been defined for this model task." )
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- llama.cpp
How to use brsx-labs/BRSX-Reasoner with llama.cpp:
Install from brew
brew install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf brsx-labs/BRSX-Reasoner # Run inference directly in the terminal: llama-cli -hf brsx-labs/BRSX-Reasoner
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf brsx-labs/BRSX-Reasoner # Run inference directly in the terminal: llama-cli -hf brsx-labs/BRSX-Reasoner
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 brsx-labs/BRSX-Reasoner # Run inference directly in the terminal: ./llama-cli -hf brsx-labs/BRSX-Reasoner
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 brsx-labs/BRSX-Reasoner # Run inference directly in the terminal: ./build/bin/llama-cli -hf brsx-labs/BRSX-Reasoner
Use Docker
docker model run hf.co/brsx-labs/BRSX-Reasoner
- LM Studio
- Jan
- Ollama
How to use brsx-labs/BRSX-Reasoner with Ollama:
ollama run hf.co/brsx-labs/BRSX-Reasoner
- Unsloth Studio new
How to use brsx-labs/BRSX-Reasoner 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 brsx-labs/BRSX-Reasoner 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 brsx-labs/BRSX-Reasoner to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for brsx-labs/BRSX-Reasoner to start chatting
- Docker Model Runner
How to use brsx-labs/BRSX-Reasoner with Docker Model Runner:
docker model run hf.co/brsx-labs/BRSX-Reasoner
- Lemonade
How to use brsx-labs/BRSX-Reasoner with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull brsx-labs/BRSX-Reasoner
Run and chat with the model
lemonade run user.BRSX-Reasoner-{{QUANT_TAG}}List all available models
lemonade list
BRSX-Reasoner-1B
A compact, reasoning-focused language model designed to operate as an internal cognition core rather than a conversational assistant.
🇹🇷 Hakkında
BRSX-Reasoner-1B, sohbet üretmek veya son kullanıcıya cevap vermek için değil,
kontrollü ve tek blokluk bir iç akıl yürütme (reasoning) üretmek üzere eğitilmiş hafif bir dil modelidir.
Bu model, daha büyük yapay zeka sistemlerinde:
- yönlendirme (routing),
- araç kullanımı (tools),
- yanıt üretimi (response models)
gibi bileşenlerle birlikte çalışacak şekilde tasarlanmış bir bilişsel çekirdek (reasoning core) olarak konumlandırılmıştır.
Bu repoda yalnızca GGUF (inference-only) dosyaları yer alır.
Modelin yeniden eğitilmesi veya ağırlıklarının değiştirilmesi önerilmez.
🇬🇧 Overview
BRSX-Reasoner-1B is a lightweight language model trained to produce a single, controlled block of internal reasoning. It is not intended to generate final user-facing answers, but to serve as a reasoning engine within modular AI architectures.
The model assumes that reasoning and response generation are separate cognitive processes and is designed accordingly.
This repository provides GGUF files for inference-only usage.
Further fine-tuning or weight modification is not recommended.
What This Model Is
- ✅ A compact reasoning-focused model
- ✅ Designed for modular AI systems
- ✅ Suitable for low-resource inference environments
- ✅ Distributed in GGUF format for stability and control
What This Model Is Not
- ❌ Not a general-purpose chatbot
- ❌ Not optimized for instruction-following or dialogue
- ❌ Not intended to produce final answers
- ❌ Not meant for further fine-tuning
Usage (Ollama)
You can integrate this model into Ollama using a custom Modelfile.
Modelfile
FROM ./.gguf
TEMPLATE """ Below is a user input followed by a single block of internal reasoning. This is not a final answer. Do not continue beyond the reasoning.
User input: {{ .Prompt }}
Reasoning: """
PARAMETER temperature 0.75 PARAMETER top_p 0.9 PARAMETER top_k 40 PARAMETER repeat_penalty 1.15 PARAMETER num_ctx 4096 PARAMETER stop "User input:" PARAMETER stop "Final answer:"
(This section is intentionally left blank.
Please define your own template and parameters based on your system design.)
Run
ollama create brsx-reasoner -f Modelfile
ollama run brsx-reasoner
> Not:
Huggingface mimariyi otomatik tespit ettiğinden dolayı gemma mimarisi yazmaktadır.
>Model gemma tabanlıdır.
Site: brsxlabs.gt.tc
- Downloads last month
- 8
We're not able to determine the quantization variants.