Instructions to use hypaai/hypa-tiny-keys with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hypaai/hypa-tiny-keys with Transformers:

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

pipe = pipeline("text-generation", model="hypaai/hypa-tiny-keys")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("hypaai/hypa-tiny-keys")
model = AutoModelForCausalLM.from_pretrained("hypaai/hypa-tiny-keys", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use hypaai/hypa-tiny-keys with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "hypaai/hypa-tiny-keys"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hypaai/hypa-tiny-keys",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/hypaai/hypa-tiny-keys

SGLang

How to use hypaai/hypa-tiny-keys 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 "hypaai/hypa-tiny-keys" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hypaai/hypa-tiny-keys",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "hypaai/hypa-tiny-keys" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hypaai/hypa-tiny-keys",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use hypaai/hypa-tiny-keys with Docker Model Runner:
```
docker model run hf.co/hypaai/hypa-tiny-keys
```

hypa-tiny-keys

This model is a fine-tuned version of on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.1086

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0003
train_batch_size: 32
eval_batch_size: 32
seed: 3407
gradient_accumulation_steps: 2
total_train_batch_size: 64
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 500
training_steps: 60000

Training results

Training Loss	Epoch	Step	Validation Loss
2.8771	0.0083	500	2.8239
2.3237	0.0167	1000	2.2951
1.9399	0.025	1500	1.9368
1.7840	0.0333	2000	1.7695
1.6826	0.0417	2500	1.6788
1.6208	0.05	3000	1.6156
1.5774	0.0583	3500	1.5753
1.5343	0.0667	4000	1.5389
1.5108	0.075	4500	1.5089
1.4918	0.0833	5000	1.4879
1.4696	0.0917	5500	1.4679
1.4439	0.1	6000	1.4490
1.4353	0.1083	6500	1.4347
1.4301	0.1167	7000	1.4197
1.4114	0.125	7500	1.4098
1.3965	0.1333	8000	1.3979
1.3879	0.1417	8500	1.3883
1.3754	0.15	9000	1.3759
1.3746	0.1583	9500	1.3687
1.3556	0.1667	10000	1.3607
1.3897	0.175	10500	1.3605
1.3735	0.1833	11000	1.3588
1.3561	0.1917	11500	1.3561
1.3514	0.2	12000	1.3529
1.3370	0.2083	12500	1.3490
1.3319	0.2167	13000	1.3470
1.3339	0.225	13500	1.3440
1.3161	0.2333	14000	1.3415
1.3142	0.2417	14500	1.3338
1.3061	0.25	15000	1.3280
1.2954	0.2583	15500	1.3279
1.2961	0.2667	16000	1.3253
1.2917	0.275	16500	1.3156
1.2850	0.2833	17000	1.3168
1.2734	0.2917	17500	1.3135
1.2637	0.3	18000	1.3080
1.2708	0.3083	18500	1.3039
1.2625	0.3167	19000	1.3002
1.2623	0.325	19500	1.2992
1.2482	0.3333	20000	1.2961
1.2453	0.3417	20500	1.2909
1.2873	0.35	21000	1.2822
1.2766	0.3583	21500	1.2847
1.2673	0.3667	22000	1.2796
1.2618	0.375	22500	1.2755
1.2543	0.3833	23000	1.2704
1.2471	0.3917	23500	1.2728
1.2345	0.4	24000	1.2703
1.2350	0.4083	24500	1.2690
1.2337	0.4167	25000	1.2657
1.2304	0.425	25500	1.2619
1.2202	0.4333	26000	1.2560
1.2182	0.4417	26500	1.2521
1.2137	0.45	27000	1.2488
1.2118	0.4583	27500	1.2478
1.2024	0.4667	28000	1.2449
1.1989	0.475	28500	1.2446
1.1929	0.4833	29000	1.2421
1.1942	0.4917	29500	1.2386
1.1842	0.5	30000	1.2354
1.1798	0.5083	30500	1.2325
1.2264	0.5167	31000	1.2270
1.2130	0.525	31500	1.2215
1.2058	0.5333	32000	1.2180
1.1970	0.5417	32500	1.2136
1.1886	0.55	33000	1.2118
1.1826	0.5583	33500	1.2106
1.1798	0.5667	34000	1.2090
1.1714	0.575	34500	1.2061
1.1707	0.5833	35000	1.2036
1.1774	0.5917	35500	1.2013
1.1602	0.6	36000	1.1985
1.1625	0.6083	36500	1.1928
1.1514	0.6167	37000	1.1921
1.1507	0.625	37500	1.1902
1.1444	0.6333	38000	1.1878
1.1425	0.6417	38500	1.1865
1.1393	0.65	39000	1.1819
1.1355	0.6583	39500	1.1793
1.1277	0.6667	40000	1.1765
1.1314	0.675	40500	1.1780
1.1272	0.6833	41000	1.1748
1.1607	0.6917	41500	1.1643
1.1553	0.7	42000	1.1613
1.1484	0.7083	42500	1.1588
1.1419	0.7167	43000	1.1546
1.1298	0.725	43500	1.1523
1.1313	0.7333	44000	1.1489
1.1322	0.7417	44500	1.1474
1.1266	0.75	45000	1.1453
1.1190	0.7583	45500	1.1429
1.1186	0.7667	46000	1.1399
1.1122	0.775	46500	1.1377
1.1145	0.7833	47000	1.1354
1.1040	0.7917	47500	1.1348
1.1126	0.8	48000	1.1331
1.1042	0.8083	48500	1.1304
1.0946	0.8167	49000	1.1298
1.0894	0.825	49500	1.1280
1.0993	0.8333	50000	1.1272
1.0916	0.8417	50500	1.1261
1.0844	0.85	51000	1.1234
1.1222	0.8583	51500	1.1204
1.1237	0.8667	52000	1.1179
1.1307	0.875	52500	1.1154
1.1195	0.8833	53000	1.1141
1.1164	0.8917	53500	1.1130
1.1139	0.9	54000	1.1124
1.1159	0.9083	54500	1.1118
1.1087	0.9167	55000	1.1112
1.1082	0.925	55500	1.1104
1.1036	0.9333	56000	1.1099
1.1099	0.9417	56500	1.1097
1.1088	0.95	57000	1.1094
1.1091	0.9583	57500	1.1091
1.1052	0.9667	58000	1.1088
1.1012	0.975	58500	1.1087
1.1072	0.9833	59000	1.1086
1.1072	0.9917	59500	1.1086
1.1053	1.0	60000	1.1086

Framework versions

Transformers 5.17.0
Pytorch 2.11.0+cu128
Datasets 5.0.1
Tokenizers 0.23.2

Downloads last month: 1,064

Safetensors

Model size

14.5M params

Tensor type

F32