Instructions to use Ayaka157/RetNet-0.5B-v21.1-chat with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Ayaka157/RetNet-0.5B-v21.1-chat with Transformers:

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

pipe = pipeline("text-generation", model="Ayaka157/RetNet-0.5B-v21.1-chat")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("Ayaka157/RetNet-0.5B-v21.1-chat", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Ayaka157/RetNet-0.5B-v21.1-chat with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Ayaka157/RetNet-0.5B-v21.1-chat"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Ayaka157/RetNet-0.5B-v21.1-chat",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Ayaka157/RetNet-0.5B-v21.1-chat

SGLang

How to use Ayaka157/RetNet-0.5B-v21.1-chat 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 "Ayaka157/RetNet-0.5B-v21.1-chat" \
    --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": "Ayaka157/RetNet-0.5B-v21.1-chat",
		"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 "Ayaka157/RetNet-0.5B-v21.1-chat" \
        --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": "Ayaka157/RetNet-0.5B-v21.1-chat",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Ayaka157/RetNet-0.5B-v21.1-chat with Docker Model Runner:
```
docker model run hf.co/Ayaka157/RetNet-0.5B-v21.1-chat
```

RetNet-0.5B v21.1-chat — served chat model

Status: usable with caveats — chat-val 2.2573, held-out 2.5898.

Repository (code, analysis, report): https://github.com/Ayaka157/RetNet
Technical report (2026-10-29): https://github.com/Ayaka157/RetNet/blob/main/docs/latex/en/RetNet-0.5B-Technical-Report.pdf

Model


Parameters	504,179,896
Architecture	RetNet (retention), 20 layers, width 1408, 22 heads, value/FFN dim 2816
Vocabulary / context	20,000 ByteLevel BPE / 2048 training context
Decay	per-head learned $\gamma$; GLA branch implemented but `gla_rank=0`
Precision	stored as fp32 safetensors (1.92 GB); load with orch_dtype=torch.bfloat16 for inference

Files

config.json, model.safetensors, modeling_retnet.py, retention_fast.py, fused_ce.py, tokenizer files. The modeling files come from the published RetNet codebase and are provided so the checkpoint can be loaded with trust_remote_code=True.

from transformers import AutoModelForCausalLM, AutoTokenizer
tok = AutoTokenizer.from_pretrained("Ayaka157/RetNet-0.5B-v21.1-chat", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("Ayaka157/RetNet-0.5B-v21.1-chat", trust_remote_code=True,
                                             torch_dtype="bfloat16")

For serving/prefill use the parallel form, training uses chunkwise, decoding uses recurrent (RETNET_FAST_CHUNKWISE=1 for batched chunkwise). The demo: https://retnet.ayaka157.icu/play/

Evidence and limitations

This is the model behind the play page. It is trained with only 0.328 B chat tokens (0.593 epoch), and the chat data is dominated by long continuations and unclosed <think> blocks. It needs an explicit concise system prompt; without one it often hits the generation cap. Local arithmetic is 0/4 and factual knowledge is weak; use it as a demo, not as a knowledge worker.

Ayaka157
/

RetNet-0.5B-v21.1-chat

RetNet-0.5B v21.1-chat — served chat model

Model

Files

Evidence and limitations

Links