Instructions to use Unseen1980/daedalus-150m-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Unseen1980/daedalus-150m-instruct with Transformers:

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

pipe = pipeline("text-generation", model="Unseen1980/daedalus-150m-instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Unseen1980/daedalus-150m-instruct")
model = AutoModelForCausalLM.from_pretrained("Unseen1980/daedalus-150m-instruct", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Unseen1980/daedalus-150m-instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Unseen1980/daedalus-150m-instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Unseen1980/daedalus-150m-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Unseen1980/daedalus-150m-instruct

SGLang

How to use Unseen1980/daedalus-150m-instruct 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 "Unseen1980/daedalus-150m-instruct" \
    --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": "Unseen1980/daedalus-150m-instruct",
		"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 "Unseen1980/daedalus-150m-instruct" \
        --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": "Unseen1980/daedalus-150m-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Unseen1980/daedalus-150m-instruct with Docker Model Runner:
```
docker model run hf.co/Unseen1980/daedalus-150m-instruct
```

Add link to paper

by nielsr HF Staff - opened Aug 25

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+13

-11

Files changed (1) hide show

README.md +13 -11

README.md CHANGED Viewed

@@ -1,18 +1,18 @@
 ---
-license: apache-2.0
 library_name: transformers
 pipeline_tag: text-generation
-language:
-  - en
 tags:
-  - daedalus
-  - cpu-inference
-  - lfm2
-  - hybrid
 widget:
-  - text: "What is the capital of France?"
-  - text: "Explain photosynthesis in one sentence."
-  - text: "What is the difference between a CPU and a GPU?"
 inference:
   parameters:
     max_new_tokens: 96
@@ -23,6 +23,8 @@ inference:
 # Daedalus-150M — Instruct
 A 150M-parameter language model built for **CPU inference**. Full attention is
 kept in only 6 of its 18 layers; the other 12 use short convolutions whose
 memory is two timesteps wide however long the conversation gets. Decoding
@@ -85,4 +87,4 @@ lines. English only, 2048-token context, single seed.
 The 4-bit build costs about 6% perplexity — quantisation-aware training was
 built but did not run. Roughly 48% of the convolution channels are inert and
 cannot be pruned, and the 49,152-entry vocabulary is larger than this model size
-warrants. All three are documented in the paper.

 ---
+language:
+- en
 library_name: transformers
+license: apache-2.0
 pipeline_tag: text-generation
 tags:
+- daedalus
+- cpu-inference
+- lfm2
+- hybrid
 widget:
+- text: What is the capital of France?
+- text: Explain photosynthesis in one sentence.
+- text: What is the difference between a CPU and a GPU?
 inference:
   parameters:
     max_new_tokens: 96
 # Daedalus-150M — Instruct
+This model was presented in the paper [Daedalus-150M: A Convolution-Attention Hybrid Designed for CPU Inference](https://huggingface.co/papers/2608.20210).
 A 150M-parameter language model built for **CPU inference**. Full attention is
 kept in only 6 of its 18 layers; the other 12 use short convolutions whose
 memory is two timesteps wide however long the conversation gets. Decoding
 The 4-bit build costs about 6% perplexity — quantisation-aware training was
 built but did not run. Roughly 48% of the convolution channels are inert and
 cannot be pruned, and the 49,152-entry vocabulary is larger than this model size
+warrants. All three are documented in the paper.