Text Generation
Transformers
Safetensors
English
Italian
quark
causal-lm
small-language-model
gqa
rope
swiglu
bash
code
custom_code
Instructions to use ThingAI/ARK-72M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ThingAI/ARK-72M with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="ThingAI/ARK-72M", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("ThingAI/ARK-72M", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use ThingAI/ARK-72M with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "ThingAI/ARK-72M" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "ThingAI/ARK-72M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/ThingAI/ARK-72M
- SGLang
How to use ThingAI/ARK-72M 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 "ThingAI/ARK-72M" \ --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": "ThingAI/ARK-72M", "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 "ThingAI/ARK-72M" \ --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": "ThingAI/ARK-72M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use ThingAI/ARK-72M with Docker Model Runner:
docker model run hf.co/ThingAI/ARK-72M
fix: inv_freq persistent=True + incluso nel safetensors
Browse files- modeling_quark.py +5 -4
modeling_quark.py
CHANGED
|
@@ -26,13 +26,14 @@ class RotaryEmbedding(nn.Module):
|
|
| 26 |
super().__init__()
|
| 27 |
assert head_dim % 2 == 0
|
| 28 |
inv_freq = 1.0 / (theta ** (torch.arange(0, head_dim, 2).float() / head_dim))
|
| 29 |
-
self.register_buffer("inv_freq", inv_freq, persistent=
|
| 30 |
self._build_cache(max_seq_len)
|
| 31 |
|
| 32 |
def _build_cache(self, seq_len):
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
|
|
|
| 36 |
self.register_buffer("cos_cache", emb.cos()[None, None], persistent=False)
|
| 37 |
self.register_buffer("sin_cache", emb.sin()[None, None], persistent=False)
|
| 38 |
self._max = seq_len
|
|
|
|
| 26 |
super().__init__()
|
| 27 |
assert head_dim % 2 == 0
|
| 28 |
inv_freq = 1.0 / (theta ** (torch.arange(0, head_dim, 2).float() / head_dim))
|
| 29 |
+
self.register_buffer("inv_freq", inv_freq, persistent=True)
|
| 30 |
self._build_cache(max_seq_len)
|
| 31 |
|
| 32 |
def _build_cache(self, seq_len):
|
| 33 |
+
device = self.inv_freq.device
|
| 34 |
+
t = torch.arange(seq_len, device=device, dtype=self.inv_freq.dtype)
|
| 35 |
+
freqs = torch.outer(t, self.inv_freq)
|
| 36 |
+
emb = torch.cat([freqs, freqs], dim=-1)
|
| 37 |
self.register_buffer("cos_cache", emb.cos()[None, None], persistent=False)
|
| 38 |
self.register_buffer("sin_cache", emb.sin()[None, None], persistent=False)
|
| 39 |
self._max = seq_len
|