Instructions to use ikaankeskin/logos-7b-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use ikaankeskin/logos-7b-gguf 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 ikaankeskin/logos-7b-gguf:Q8_0
# Run inference directly in the terminal:
llama cli -hf ikaankeskin/logos-7b-gguf:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf ikaankeskin/logos-7b-gguf:Q8_0
# Run inference directly in the terminal:
llama cli -hf ikaankeskin/logos-7b-gguf:Q8_0

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 ikaankeskin/logos-7b-gguf:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf ikaankeskin/logos-7b-gguf:Q8_0

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 ikaankeskin/logos-7b-gguf:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf ikaankeskin/logos-7b-gguf:Q8_0

Use Docker

docker model run hf.co/ikaankeskin/logos-7b-gguf:Q8_0

LM Studio
Jan

vLLM

How to use ikaankeskin/logos-7b-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ikaankeskin/logos-7b-gguf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ikaankeskin/logos-7b-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ikaankeskin/logos-7b-gguf:Q8_0

Ollama
How to use ikaankeskin/logos-7b-gguf with Ollama:
```
ollama run hf.co/ikaankeskin/logos-7b-gguf:Q8_0
```
Unsloth Desktop
Docker Model Runner
How to use ikaankeskin/logos-7b-gguf with Docker Model Runner:
```
docker model run hf.co/ikaankeskin/logos-7b-gguf:Q8_0
```

Lemonade

How to use ikaankeskin/logos-7b-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ikaankeskin/logos-7b-gguf:Q8_0

Run and chat with the model

lemonade run user.logos-7b-gguf-Q8_0

List all available models

lemonade list

Atomic Chat

LoGos-7B GGUF (Q8_0)

Local GGUF conversion of YichuanMa/LoGos-7B for Ollama and llama.cpp. Used by Kaibitzer.

This repo is not a new training run. Weights come from the original model; they are stored as Q8_0 (8-bit) GGUF so they fit on a consumer GPU. Credit, paper, and Apache-2.0 license follow the original authors.

What is LoGos?

LoGos-7B is a 7B language model for Go (weiqi) reasoning: it reads a board position, thinks in a long chain-of-thought, and proposes the next move (color, coordinate, and a win-rate estimate).

It is built on Qwen2.5-7B, then mixed cold-start training plus GRPO so professional Go knowledge and long-CoT reasoning transfer onto actual games. It is a tutor / analysis model, not a replacement for a full-strength engine like KataGo.

Paper: Mixing Expert Knowledge: Bring Human Thoughts Back To the Game of Go (Ma et al., 2026).

Original weights are BF16 safetensors (~~14 GB). This file is the same model in Q8_0 GGUF (~~8.1 GB).

Quantization

File	Format	Quant	Size	vs original
`logos-7b-q8_0.gguf`	GGUF	Q8_0 (8-bit)	~8.1 GB	Converted from BF16; small quality drop vs the Hub safetensors, much smaller than a 4-bit quant

Yes, it is quantized. Say so when you share the file: Hub listings, Ollama, and llama.cpp all treat Q8_0 as a distinct artifact from the original BF16.

Q8_0 is a near-lossless 8-bit quant (typical choice when you want local inference without a heavy quality hit). It is not Q4. A Q4_K_M built by requantizing this Q8 file would lose more quality; a proper smaller quant should be converted from higher precision later.

Verified in Kaibitzer via Ollama on an RTX 3080 Laptop (~8 GB VRAM, ~2.3 tok/s on long CoT). Generation is slow at Q8; that is expected.

Also on the Ollama library: ikaankeskin/logos-7b.

Run it

ollama pull ikaankeskin/logos-7b

# or from this repo:
ollama pull hf.co/ikaankeskin/logos-7b-gguf:logos-7b-q8_0.gguf
ollama cp hf.co/ikaankeskin/logos-7b-gguf:logos-7b-q8_0.gguf logos-7b

Kaibitzer defaults: LOGOS_URL=http://127.0.0.1:11434, LOGOS_MODEL=logos-7b. For the Chrome/web client, set OLLAMA_ORIGINS=*.

Conversion

From the original Hugging Face safetensors, llama.cpp convert_hf_to_gguf.py → Q8_0. Direct Ollama safetensors import of Qwen2ForCausalLM failed on Ollama 0.18 (unsupported architecture); GGUF is the working path.

Kaibitzer uses context 4096 and num_predict 512.

Prompt

LoGos expects its Chinese training template: move record (1.X-Q16), board matrix (1 black, -1 white, 0 empty), then boxed 下一步位置. Kaibitzer builds that prompt for you. See the original model card for the full template.

License

Apache-2.0, same as YichuanMa/LoGos-7B.

@misc{ma2026mixingexpertknowledgebring,
      title={Mixing Expert Knowledge: Bring Human Thoughts Back To the Game of Go},
      author={Yichuan Ma and Linyang Li and Yongkang Chen and Peiji Li and Jiasheng Ye and Qipeng Guo and Dahua Lin and Kai Chen},
      year={2026},
      eprint={2601.16447},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2601.16447},
}