Instructions to use Pinkstack/fijik-2.0-350m-sft-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Pinkstack/fijik-2.0-350m-sft-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Pinkstack/fijik-2.0-350m-sft-GGUF",
	filename="Fijik-2.0-350M-SFT-bf16.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Pinkstack/fijik-2.0-350m-sft-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 Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M

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 Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M

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 Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M

Use Docker

docker model run hf.co/Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use Pinkstack/fijik-2.0-350m-sft-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Pinkstack/fijik-2.0-350m-sft-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": "Pinkstack/fijik-2.0-350m-sft-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M

Ollama
How to use Pinkstack/fijik-2.0-350m-sft-GGUF with Ollama:
```
ollama run hf.co/Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M
```

Unsloth Studio

How to use Pinkstack/fijik-2.0-350m-sft-GGUF 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 Pinkstack/fijik-2.0-350m-sft-GGUF 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 Pinkstack/fijik-2.0-350m-sft-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Pinkstack/fijik-2.0-350m-sft-GGUF to start chatting

How to use Pinkstack/fijik-2.0-350m-sft-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Pinkstack/fijik-2.0-350m-sft-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M

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 Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use Pinkstack/fijik-2.0-350m-sft-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M

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 "Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M" \
  --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"

Docker Model Runner
How to use Pinkstack/fijik-2.0-350m-sft-GGUF with Docker Model Runner:
```
docker model run hf.co/Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M
```

Lemonade

How to use Pinkstack/fijik-2.0-350m-sft-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Pinkstack/fijik-2.0-350m-sft-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.fijik-2.0-350m-sft-GGUF-Q4_K_M

List all available models

lemonade list

(gguf version - correct ggufs have been uploaded)

Fijik 2.0 350M SFT

Fijik 2.0 350M SFT is the first, agentic, edge-LLM we have ever released. it is designed specifically for edge devices and quick inference it was trained efficiently on just one GPU. In addition the model can work with up to 65K context (thanks to YARN scaling) and is based off of granite-4.0-350M-base. (non H, aka no mamba layers).

Fijik 2.0 supports the following reasoning efforts, configurable via the chat template: disabled <--- Configures the model to not think at all, it will not generate reasoning tags, best for chat, title generation.

low <--- Configures the model to think a little before responding, good for web search, tool calls etc.

medium <--- Configures the model to think well before responding, good for code etc.

high <--- Configures the model to think hard before responding, may loop, could improve math, code performance, not needed for most users.

We recommend the following sampling parameters:

tempature: 0.7
Top K: 40
Repeat Penalty: 1.05
Presence Penalty: 0.25
Top P Sampling: 0.95
Min P sampling: 0 or disabled

Benchmarks

NOTE: all are pass@4, aka "could the model get it right at least once"

NOTE2: all benchmarks were run internally and using the exact same sampling parameters

Benchmark	Fijik2.0-350M-d	LFM2.5-350M	granite4.0-h-350M	SmolLM2-360M
MATH500	9.20	21.80	29.60	1.40**
GPQA-DIAMOND	65.15	33.33	53.03	errors**
MMLU REDUX*	65.60	47.20	61.07	nill
HUMANEVAL	22.56	17.68	38.41	9.76

For fijik2.0: d efforts: d = disabled to make benchmark(s) fair as other than fijik all the others are non-reasoning models

*mmlu_redux only 10 subjects, 750 items.

**SmolLM2 often fell into loops, not enough context, etc. MMLU not included, math500 had 7% errors and thus is not trustworthy. GPQA had so many errors that it has been removed for smollm2.

In this benchmark table we can see Fijik2.0 performing competitively with similarly sized models, though fijik2.0 is weak on math, probably due to no RL.

Training metholodgy

    Granite 4.0 350M base
      \           /
    Continual pre-training1 (~6B tokens, august 2025 knowledge cutoff, low context)
       \         / 
    Continual pre-training2 (~1B tokens, aug 2025 knowledge cutoff, higher context)
        \       /
    Supervised fine-tuning  (~3B tokens, data from feb 2026, mixed chat, agentic, web search, code, tool use, reasoning)
          \   /
    Final uploaded model    (best checkpoint chosen)

For CPT a 2bit ademamix-style optimizer was used. For SFT a 4bit ademamix-style optimizer was used. Both optimizers worked well. Fijik 2.0 got 1.1 loss at its selected sft checkpoint. (1.3 EPOCH)

Safety

Fijik-2.0 has not undergone RLHF. Although it still is safety trained thanks to SFT. In addition fijik has been trained to state when it does not know something. Though ideally in the system prompt tell it specifically that if it has no information about something, then it should state so.

Compute

1x rtx 2000 ada PCIE (16gb, overclocked, 70 watt limit)

Thanks

Thank you for your support 💕

Downloads last month: 223

GGUF

Model size

0.4B params

Architecture

granite

Hardware compatibility

4-bit

5-bit

6-bit

8-bit

16-bit

Model tree for Pinkstack/fijik-2.0-350m-sft-GGUF

Base model

Pinkstack/fijik-2.0-350m-sft

Quantized

(1)

this model

Collection including Pinkstack/fijik-2.0-350m-sft-GGUF

Fijik 2.0

Collection

Small yet agentic with an open license • 2 items • Updated 1 day ago