Instructions to use ASTRAI-labs/pluto-nano-0.5-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ASTRAI-labs/pluto-nano-0.5-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="ASTRAI-labs/pluto-nano-0.5-gguf",
	filename="pluto-nano-0.5.fp16.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use ASTRAI-labs/pluto-nano-0.5-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 ASTRAI-labs/pluto-nano-0.5-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ASTRAI-labs/pluto-nano-0.5-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 ASTRAI-labs/pluto-nano-0.5-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ASTRAI-labs/pluto-nano-0.5-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 ASTRAI-labs/pluto-nano-0.5-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf ASTRAI-labs/pluto-nano-0.5-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 ASTRAI-labs/pluto-nano-0.5-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf ASTRAI-labs/pluto-nano-0.5-gguf:Q4_K_M

Use Docker

docker model run hf.co/ASTRAI-labs/pluto-nano-0.5-gguf:Q4_K_M

LM Studio
Jan

vLLM

How to use ASTRAI-labs/pluto-nano-0.5-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ASTRAI-labs/pluto-nano-0.5-gguf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ASTRAI-labs/pluto-nano-0.5-gguf",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ASTRAI-labs/pluto-nano-0.5-gguf:Q4_K_M

Ollama
How to use ASTRAI-labs/pluto-nano-0.5-gguf with Ollama:
```
ollama run hf.co/ASTRAI-labs/pluto-nano-0.5-gguf:Q4_K_M
```

Unsloth Studio

How to use ASTRAI-labs/pluto-nano-0.5-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 ASTRAI-labs/pluto-nano-0.5-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 ASTRAI-labs/pluto-nano-0.5-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for ASTRAI-labs/pluto-nano-0.5-gguf to start chatting

Atomic Chat new
Docker Model Runner
How to use ASTRAI-labs/pluto-nano-0.5-gguf with Docker Model Runner:
```
docker model run hf.co/ASTRAI-labs/pluto-nano-0.5-gguf:Q4_K_M
```

Lemonade

How to use ASTRAI-labs/pluto-nano-0.5-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ASTRAI-labs/pluto-nano-0.5-gguf:Q4_K_M

Run and chat with the model

lemonade run user.pluto-nano-0.5-gguf-Q4_K_M

List all available models

lemonade list

ASTRAI Pluto Nano 0.5 — GGUF

Quantized GGUF builds of ASTRAI-labs/pluto-nano-0.5 for use with llama.cpp (and downstream wrappers: Ollama, LM Studio, etc.).

⚠️ Unofficial shim

Pluto Nano's native architecture (astrai_pluto) is not yet supported upstream in llama.cpp. These files were produced by re-mapping Pluto's tensors onto the qwen2_moe architecture with dummy zero-weight shared-expert tensors (Pluto uses top-1 routing without a shared expert). See the conversion script & rationale in GGUF.md.

Outputs match the bf16 safetensors version. Quality at low quantization (Q4_K_M) may degrade slightly because top-1 MoE has higher per-token sensitivity than the top-4 MoE that qwen2_moe quantizers assume.

Files

File	Size	Bits	Use case
`pluto-nano-0.5.fp16.gguf`	2.0 GB	16-bit	reference / fine-tune base
`pluto-nano-0.5.q8_0.gguf`	1.1 GB	8.5-bit	≈ FP8, near-zero quality loss
`pluto-nano-0.5.q6_K.gguf`	963 MB	6.6-bit	small + clean
`pluto-nano-0.5.q4_K_M.gguf`	679 MB	4.8-bit	smallest practical

Usage

# Download
hf download ASTRAI-labs/pluto-nano-0.5-gguf pluto-nano-0.5.q8_0.gguf

# Run
./llama-cli -m pluto-nano-0.5.q8_0.gguf \
  --prompt "<|lang_en|>\n<|user|>\nHi, who are you?\n<|im_end|>\n<|assistant|>\n" \
  -n 200 --temp 0.7

Chat template

<|lang_{en|pt|es|zh|hi}|>
<|user|>
...question...
<|im_end|>
<|assistant|>
...response...
<|im_end|>

License

ASTRAI Closed License — see main repo.

Downloads last month: 213

GGUF

Model size

1B params

Architecture

qwen2moe

Hardware compatibility

4-bit

6-bit

8-bit

View +1 variant

Model tree for ASTRAI-labs/pluto-nano-0.5-gguf

Base model

ASTRAI-labs/pluto-nano-0.5-base

Finetuned

ASTRAI-labs/pluto-nano-0.5

Quantized

(1)

this model