Instructions to use RusMiniLM/RusNanoLM-800M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use RusMiniLM/RusNanoLM-800M with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="RusMiniLM/RusNanoLM-800M") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("RusMiniLM/RusNanoLM-800M") model = AutoModelForMultimodalLM.from_pretrained("RusMiniLM/RusNanoLM-800M", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use RusMiniLM/RusNanoLM-800M 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 RusMiniLM/RusNanoLM-800M:F16 # Run inference directly in the terminal: llama cli -hf RusMiniLM/RusNanoLM-800M:F16
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf RusMiniLM/RusNanoLM-800M:F16 # Run inference directly in the terminal: llama cli -hf RusMiniLM/RusNanoLM-800M:F16
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 RusMiniLM/RusNanoLM-800M:F16 # Run inference directly in the terminal: ./build/bin/llama cli -hf RusMiniLM/RusNanoLM-800M:F16
Use Docker
docker model run hf.co/RusMiniLM/RusNanoLM-800M:F16
- Llama new
- LM Studio
- Jan
- vLLM
How to use RusMiniLM/RusNanoLM-800M with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "RusMiniLM/RusNanoLM-800M" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "RusMiniLM/RusNanoLM-800M", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/RusMiniLM/RusNanoLM-800M:F16
- SGLang
How to use RusMiniLM/RusNanoLM-800M 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 "RusMiniLM/RusNanoLM-800M" \ --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": "RusMiniLM/RusNanoLM-800M", "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 "RusMiniLM/RusNanoLM-800M" \ --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": "RusMiniLM/RusNanoLM-800M", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Ollama
How to use RusMiniLM/RusNanoLM-800M with Ollama:
ollama run hf.co/RusMiniLM/RusNanoLM-800M:F16
- Unsloth Desktop
- Pi
How to use RusMiniLM/RusNanoLM-800M with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf RusMiniLM/RusNanoLM-800M:F16
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:9931/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "RusMiniLM/RusNanoLM-800M:F16" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use RusMiniLM/RusNanoLM-800M with Docker Model Runner:
docker model run hf.co/RusMiniLM/RusNanoLM-800M:F16
- Lemonade
How to use RusMiniLM/RusNanoLM-800M with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull RusMiniLM/RusNanoLM-800M:F16
Run and chat with the model
lemonade run user.RusNanoLM-800M-F16
List all available models
lemonade list
- Hermes Agent
How to use RusMiniLM/RusNanoLM-800M with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf RusMiniLM/RusNanoLM-800M:F16
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:9931/v1 hermes config set model.default RusMiniLM/RusNanoLM-800M:F16
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use RusMiniLM/RusNanoLM-800M with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf RusMiniLM/RusNanoLM-800M:F16
Configure OpenClaw
# Install OpenClaw: npm install -g openclaw@latest # Register the local server and set it as the default model: openclaw onboard --non-interactive --mode local \ --auth-choice custom-api-key \ --custom-base-url http://127.0.0.1:9931/v1 \ --custom-model-id "RusMiniLM/RusNanoLM-800M:F16" \ --custom-provider-id llama-cpp \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
RusNanoLM-800M
English Overview
RusNanoLM-800M is an open-weights 800M parameter bilingual (Russian / English) language model built on a hybrid Linear Attention architecture (Qwen3.5 with Gated Delta Rule layers). It is evaluated on MERA, the Russian language understanding benchmark suite: on RuMultiAR a 0.8B model outperforms Qwen2.5-7B and GigaChat-2-10B.
MERA Results (Pass@1, greedy, chat template)
| Benchmark | Tasks | RusNanoLM-0.8B | Qwen3.5-0.8B (baseline) | ฮ |
|---|---|---|---|---|
| SimpleAR | 1000 | 83.2% | 48.3% | +34.9 |
| RuMultiAR | 1039 | 33.5% | 8.1% | +25.4 (ร4.1) |
| MathLogicQA | 680 | 38.2% | 33.9% | +4.4 |
| MaMuRaMu | 285 | 46.0% | 43.6% | +2.4 |
| PARus | 500 | 53.4% | 53.6% | -0.2 |
| RWSD | 810 | 49.9% | 52.7% | -2.8 |
| Strict Constraints JSON | 12 | 58.3% | - | - |
Evaluation date: 2026-11-17. Baseline numbers โ official MERA submission of Qwen3.5-0.8B (submit id 16125).
RuMultiAR vs larger models
| Model | Params | RuMultiAR |
|---|---|---|
| RuadaptQwen-32B | 32B | 43.3% |
| RusNanoLM-0.8B (ours) | 0.8B | 33.5% |
| GigaChat-2-10B | 10B | 31.5% |
| Qwen2.5-7B | 7B | 30.8% |
| Meta-Llama-3-8B | 8B | 24.8% |
| Llama-3.2-1B-Instruct | 1B | 16.2% |
| Qwen3.5-0.8B (baseline) | 0.8B | 8.1% |
Visual Leaderboards
MERA Suite Summary
Baseline Progression
SimpleAR
RuMultiAR
MathLogicQA
MaMuRaMu
Quickstart & Inference
import torch
from transformers import AutoModelForImageTextToText, AutoTokenizer
MODEL_ID = "itismax0/RusNanoLM-800M"
tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
model = AutoModelForImageTextToText.from_pretrained(
MODEL_ID,
torch_dtype=torch.bfloat16,
device_map="auto"
)
model.eval()
messages = [
{"role": "user", "content": "ะ ะตัะธ ะทะฐะดะฐัั: ะฒ ะบะพัะทะธะฝะต 5 ัะฑะปะพะบ ะธ 7 ะณััั. ะกะบะพะปัะบะพ ะฒัะตะณะพ ัััะบัะพะฒ?"}
]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
with torch.no_grad():
outputs = model.generate(
**inputs,
max_new_tokens=512,
do_sample=False,
pad_token_id=tokenizer.eos_token_id
)
response = tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True)
print(response)
Run with llama.cpp / Ollama (GGUF, full precision)
Two full-precision F16 GGUF files are included โ no quantization loss. They differ only in tensor naming (llama.cpp renamed some Qwen3.5 tensors between b12257 and b12413):
| File | Use with |
|---|---|
RusNanoLM-800M-F16.gguf |
llama.cpp >= b12413 (current master) |
RusNanoLM-800M-F16-ollama.gguf |
Ollama (all current versions) and other apps bundling llama.cpp <= b12257 (LM Studio, etc.) |
# llama.cpp (fresh build)
llama-cli -hf itismax0/RusNanoLM-800M:RusNanoLM-800M-F16.gguf --temp 0 -p "ะ ะตัะธ: 13*7"
# Ollama โ direct pull
ollama run hf.co/itismax0/RusNanoLM-800M:RusNanoLM-800M-F16-ollama.gguf "ะััะธัะปะธ: 13*7"
Note: the model uses the Qwen3.5 hybrid linear-attention architecture. If you see
tensor 'blk.0.ssm_beta.weight' not foundin Ollama โ use the-ollamaGGUF variant. If LM Studio refuses to load, update its llama.cpp runtime (Settings โ Runtimes).
ะะฑะทะพั ะฝะฐ ััััะบะพะผ
RusNanoLM-800M โ ะพัะบัััะฐั ะดะฒััะทััะฝะฐั (ััััะบะธะน ะธ ะฐะฝะณะปะธะนัะบะธะน) ะผะพะดะตะปั ะฝะฐ 800 ะผะปะฝ ะฟะฐัะฐะผะตััะพะฒ, ะฟะพัััะพะตะฝะฝะฐั ะฝะฐ ะณะธะฑัะธะดะฝะพะน ะฐัั ะธัะตะบัััะต Qwen3.5 ั ะปะธะฝะตะนะฝัะผ ะฒะฝะธะผะฐะฝะธะตะผ (Gated Delta Rule): 24 ัะปะพั, ะบะฐะถะดัะน ัะตัะฒััััะน โ ะฟะพะปะฝะพะต ะฒะฝะธะผะฐะฝะธะต. ะัะตะฝะธะฒะฐะปะฐัั ะฝะฐ ััััะบะพัะทััะฝะพะผ ะฑะตะฝัะผะฐัะบะต MERA (Pass@1, ะถะฐะดะฝะฐั ะณะตะฝะตัะฐัะธั, chat template).
ะ ะตะทัะปััะฐัั MERA
| ะะตะฝัะผะฐัะบ | ะะฐะดะฐั | RusNanoLM-0.8B | Qwen3.5-0.8B (baseline) | ฮ |
|---|---|---|---|---|
| SimpleAR | 1000 | 83.2% | 48.3% | +34.9 |
| RuMultiAR | 1039 | 33.5% | 8.1% | +25.4 (ร4.1) |
| MathLogicQA | 680 | 38.2% | 33.9% | +4.4 |
| MaMuRaMu | 285 | 46.0% | 43.6% | +2.4 |
| PARus | 500 | 53.4% | 53.6% | -0.2 |
| RWSD | 810 | 49.9% | 52.7% | -2.8 |
| Strict Constraints JSON | 12 | 58.3% | - | - |
ะะปะฐะฒะฝัะน ัะตะทัะปััะฐั: RuMultiAR ะฟัะพัะธะฒ ะฑะพะปััะธั ะผะพะดะตะปะตะน
ะะฐ ะฐัะธัะผะตัะธัะตัะบะธั ัะฐัััะถะดะตะฝะธัั (RuMultiAR) ะบะพะผะฟะฐะบัะฝะฐั ะผะพะดะตะปั 0.8B ะพะฑะณะพะฝัะตั ะผะพะดะตะปะธ ะฒ 8-12 ัะฐะท ััะถะตะปะตะต:
- 33.5% โ RusNanoLM-0.8B (ะฝะฐัะฐ)
- 31.5% โ GigaChat-2-10B
- 30.8% โ Qwen2.5-7B
- 24.8% โ Meta-Llama-3-8B
- 8.1% โ ะฑะฐะทะพะฒะฐั Qwen3.5-0.8B (ัะพัั ะฒ 4.1 ัะฐะทะฐ)
ะงะตััะฝัะต ะพะณัะฐะฝะธัะตะฝะธั
- PARus (-0.2%) ะธ RWSD (-2.8%) โ ะฝะฐ ััะพะฒะฝะต ะธะปะธ ัััั ะฝะธะถะต baseline: ะฟัะธัะพัั ะฟะพะปััะตะฝ ะทะฐ ัััั ะฐัะธัะผะตัะธะบะธ ะธ ัะฐัััะถะดะตะฝะธะน, ะพะฑัะตะต ะฟะพะฝะธะผะฐะฝะธะต ัะตะบััะฐ ะฝะต ะดะตะณัะฐะดะธัะพะฒะฐะปะพ, ะฝะพ ะธ ะฝะต ะฒััะพัะปะพ.
- Strict Constraints JSON โ ะฒัะตะณะพ 12 ะทะฐะดะฐั, ะฒัะฑะพัะบะฐ ะผะฐะปะฐ ะดะปั ัะธะปัะฝัั ะฒัะฒะพะดะพะฒ.
ะัะฐัะธะบะธ
ะัั ะธัะตะบัััะฐ ะธ ะฟะฐัะฐะผะตััั
- ะะฐัะฐะผะตััั: 800 ะผะปะฝ (0.8B)
- ะะฝะธะผะฐะฝะธะต: ะณะธะฑัะธะดะฝะพะต ะปะธะฝะตะนะฝะพะต ะฒะฝะธะผะฐะฝะธะต (Gated Delta Rule), ัะปะพะถะฝะพััั ะฟะฐะผััะธ ัะพััะพัะฝะธั O(1)
- ะกะปะพะธ: 24, ะบะฐะถะดัะน ัะตัะฒััััะน โ ะฟะพะปะฝะพะต ะฒะฝะธะผะฐะฝะธะต
- ะขัะตะฑะพะฒะฐะฝะธั ะบ ะฒะธะดะตะพะฟะฐะผััะธ: ะพะบะพะปะพ 1.7 ะะ ะฒ ัะพัะผะฐัะต BF16 (ะทะฐะฟััะบะฐะตััั ะฝะฐ GPU ั 4 ะะ VRAM)
- ะจะฐะฑะปะพะฝ ะดะธะฐะปะพะณะฐ: ChatML (ะฒัััะพะตะฝ ะฒ tokenizer_config)
- ะะธัะตะฝะทะธั: Apache 2.0
- Downloads last month
- 194





