Instructions to use RobTeam/rob-1-flash-2b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use RobTeam/rob-1-flash-2b 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 RobTeam/rob-1-flash-2b:F16
# Run inference directly in the terminal:
llama cli -hf RobTeam/rob-1-flash-2b:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf RobTeam/rob-1-flash-2b:F16
# Run inference directly in the terminal:
llama cli -hf RobTeam/rob-1-flash-2b:F16

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 RobTeam/rob-1-flash-2b:F16
# Run inference directly in the terminal:
./llama-cli -hf RobTeam/rob-1-flash-2b:F16

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 RobTeam/rob-1-flash-2b:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf RobTeam/rob-1-flash-2b:F16

Use Docker

docker model run hf.co/RobTeam/rob-1-flash-2b:F16

LM Studio
Jan
Ollama
How to use RobTeam/rob-1-flash-2b with Ollama:
```
ollama run hf.co/RobTeam/rob-1-flash-2b:F16
```
Unsloth Desktop

How to use RobTeam/rob-1-flash-2b with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RobTeam/rob-1-flash-2b: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:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "RobTeam/rob-1-flash-2b:F16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use RobTeam/rob-1-flash-2b with Docker Model Runner:
```
docker model run hf.co/RobTeam/rob-1-flash-2b:F16
```

Lemonade

How to use RobTeam/rob-1-flash-2b with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull RobTeam/rob-1-flash-2b:F16

Run and chat with the model

lemonade run user.rob-1-flash-2b-F16

List all available models

lemonade list

Hermes Agent

How to use RobTeam/rob-1-flash-2b with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RobTeam/rob-1-flash-2b: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://127.0.0.1:8080/v1
hermes config set model.default RobTeam/rob-1-flash-2b:F16

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use RobTeam/rob-1-flash-2b with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RobTeam/rob-1-flash-2b: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://127.0.0.1:8080/v1 \
  --custom-model-id "RobTeam/rob-1-flash-2b: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"

rob-1-flash-2b (GGUF) — unofficial derivative

Russian chat model of the rob-1 line (flash tier). Built on Qwen/Qwen3.5-2B (Apache 2.0), not affiliated with Alibaba Qwen team. Base license applies.

What was done

SFT (LoRA r=8, attention + MLP, base frozen) on 2934 clean Russian rows: everyday Q&A, math, Python/Java/Lua/C++ code, identity rows (rob-1-flash, creator robanik). Shuffled, CPU.
Adapters merged into stock weights (MTP tensors preserved).
GGUF via llama.cpp: F16 (3.9 GB) + Q8_0 (1.9 GB).
Engine-verified in llama-server: math, Russian small talk, identity (Меня зовут rob-1-flash), ~15 tok/s CPU.

Files

rob-1-flash-2b-Q8_0.gguf (1.9 GB) — recommended for LM Studio / Ollama.
rob-1-flash-2b-F16.gguf (3.9 GB) — full precision merged.

Use (LM Studio)

Pull this repo from the Models tab or drop a .gguf in, context 4096. Sampling: temperature 1.0, top_k 20, top_p 1.0, presence_penalty 2.0.

Honesty notes

No formal benchmarks were run; domain spot-checks only.
Small SFT shifts style toward the training rows; hard reasoning stays at base 2B level. No scripts or training code distributed.

Downloads last month: 3

GGUF

Model size

2B params

Architecture

qwen35

Hardware compatibility

8-bit

16-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for RobTeam/rob-1-flash-2b

Base model

Qwen/Qwen3.5-2B-Base

Finetuned

Qwen/Qwen3.5-2B

Quantized

(215)

this model