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 Rahspide/MNN-Pocket-0.3B-RU-Chat-GGUF:
# Run inference directly in the terminal:
llama cli -hf Rahspide/MNN-Pocket-0.3B-RU-Chat-GGUF:
Install from WinGet (Windows)
winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Rahspide/MNN-Pocket-0.3B-RU-Chat-GGUF:
# Run inference directly in the terminal:
llama cli -hf Rahspide/MNN-Pocket-0.3B-RU-Chat-GGUF:
Build from source code
git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf Rahspide/MNN-Pocket-0.3B-RU-Chat-GGUF:
# Run inference directly in the terminal:
./build/bin/llama cli -hf Rahspide/MNN-Pocket-0.3B-RU-Chat-GGUF:
Use Docker
docker model run hf.co/Rahspide/MNN-Pocket-0.3B-RU-Chat-GGUF:
Quick Links

MNN-Pocket-0.3B-RU-Chat GGUF

A Russian instruction-following assistant in 0.36 B parameters, converted to GGUF for llama.cpp. It answers instead of continuing. 50 M tokens of SFT on top of the continuation model produced that.

Weights, data and evaluation live in Rahspide/MNN-Pocket-0.3B-RU-Chat.

Run

The ChatML template is inside the file, so --jinja is enough:

llama-cli -hf Rahspide/MNN-Pocket-0.3B-RU-Chat-GGUF:Q4_K_M --jinja \
  -p "ะžะฑัŠััะฝะธ, ั‡ั‚ะพ ั‚ะฐะบะพะต ะธะฝั„ะปัั†ะธั" -n 60

Without the template the model continues text instead of answering and behaves like its sister.

Quants

File Size Note
F16 581 MB reference, not for shipping
Q8_0 310 MB near-lossless, desktop
Q4_K_M 223 MB phone size
Q4_1 208 MB fastest generation on CPU

Measured on a 16-thread x86 CPU: ~123 tok/s generation on Q4_K_M, ~102 tok/s on Q8_0.

Measured back to back on the same 16-thread x86 CPU: Q4_1 generates 62.6 tok/s against 54.3 for Q4_K_M, at the cost of prompt processing (199 vs 251 tok/s).

Downloads last month
350
GGUF
Model size
0.3B params
Architecture
qwen3
Hardware compatibility
Log In to add your hardware

4-bit

8-bit

16-bit

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

Model tree for Rahspide/MNN-Pocket-0.3B-RU-Chat-GGUF