Instructions to use igorktech/nanofly-decoder-en with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use igorktech/nanofly-decoder-en with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="igorktech/nanofly-decoder-en", trust_remote_code=True)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("igorktech/nanofly-decoder-en", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use igorktech/nanofly-decoder-en with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/igorktech/nanofly-decoder-en

SGLang

How to use igorktech/nanofly-decoder-en 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 "igorktech/nanofly-decoder-en" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "igorktech/nanofly-decoder-en",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "igorktech/nanofly-decoder-en" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "igorktech/nanofly-decoder-en",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use igorktech/nanofly-decoder-en with Docker Model Runner:
```
docker model run hf.co/igorktech/nanofly-decoder-en
```

nanofly-decoder-en / config.json

igorktech

nanoFLY decoder: MaleCNS v1.0 connectome as a frozen recurrent layer

23cc7a6 17 days ago

Raw History Blame Contribute Delete

1.09 kB

	{
	"arch": "decoder",
	"architectures": [
	"FlyForCausalLM"
	],
	"auto_map": {
	"AutoConfig": "configuration_fly.FlyConfig",
	"AutoModel": "modeling_fly.FlyModel",
	"AutoModelForCausalLM": "modeling_fly.FlyForCausalLM"
	},
	"bos_token_id": 1,
	"connectome": "MaleCNS v1.0, FlyEM / HHMI Janelia, University of Cambridge, MRC LMB, Google Research. CC BY 4.0.",
	"d_emb": 256,
	"delay": 8,
	"dtype": "float32",
	"eos_token_id": 2,
	"head_type": "lowrank",
	"min_syn": 1,
	"mode": "gains",
	"model_type": "fly",
	"modulatory_sign": 0.0,
	"n_edges": 9055280,
	"n_neurons": 49393,
	"n_news_input": 0,
	"n_reserved": 2635,
	"n_token_input": 11434,
	"news_dim": 0,
	"news_encoder": "none",
	"news_glom": 0,
	"news_group": "orn",
	"news_mode": "glomeruli",
	"news_prefix": "",
	"pad_token_id": 0,
	"readout": "all",
	"readout_rank": 256,
	"readout_size": 49393,
	"source_repo": "",
	"ticks": 2,
	"tie_word_embeddings": false,
	"token_input": "cb_sensory,visual_projection",
	"transformers_version": "5.17.0",
	"use_cache": true,
	"vocab_size": 1024
	}