Instructions to use abcsk123/PyraCode-1.5B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use abcsk123/PyraCode-1.5B with Transformers:

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

pipe = pipeline("text-generation", model="abcsk123/PyraCode-1.5B", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("abcsk123/PyraCode-1.5B", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use abcsk123/PyraCode-1.5B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "abcsk123/PyraCode-1.5B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "abcsk123/PyraCode-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/abcsk123/PyraCode-1.5B

SGLang

How to use abcsk123/PyraCode-1.5B 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 "abcsk123/PyraCode-1.5B" \
    --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": "abcsk123/PyraCode-1.5B",
		"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 "abcsk123/PyraCode-1.5B" \
        --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": "abcsk123/PyraCode-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use abcsk123/PyraCode-1.5B with Docker Model Runner:
```
docker model run hf.co/abcsk123/PyraCode-1.5B
```

PyraCode-1.5B / README.md

abcsk123

Update README.md

76f5fd3 verified 3 months ago

preview code

Raw

History Blame Contribute Delete

1.85 kB

	---
	license: mit
	datasets:
	- tokyotech-llm/swallow-code-v2
	language:
	- en
	- zh
	metrics:
	- accuracy
	base_model:
	- Qwen/Qwen2.5-Coder-1.5B
	library_name: transformers
	tags:
	- Qwen
	- HybridArch
	- sinkAttention
	- MLA
	- GQA
	---


	# PyraCode-1.5B

	## 🌟 Model Overview
	This is a custom-architected model based on `Qwen2.5-Coder-1.5B`. We introduced a novel Asymmetric Hybrid Architecture (GQA + MLA) with Cross-Layer Shared Latent Gates and Attention Sinks, enabling efficient feature communication and reduced KV-Cache memory footprint.

	## 🏗️ Architecture Innovations

	![image](https://cdn-uploads.huggingface.co/production/uploads/67cd51087c6e6ea1cc18d236/-rJ7l5MLkX0orahfKVl3L.png)

	Unlike standard Qwen2 models, this `Hybrid-v9` backbone features:
	1. Asymmetric Layers:
	* L0-L6: Standard GQA (Grouped-Query Attention) for robust low-level feature extraction.
	* L7 (Shared Hub): Generates a global latent vector $c_{kv}$ (Rank 320).
	* L8-L27: Soft MLA (Multi-Head Latent Attention) with SVD-initialized low-rank projections.
	2. Shared Latent Gate: Deep layers can dynamically access the global latent vector from L7 via a learnable gating mechanism (`warmup_alpha`).
	3. HybridCache & Attention Sinks: Implements a sliding window (8192) alongside a 64-token attention sink to maintain generation stability at infinite sequence lengths.

	## 🚀 Quick Start

	⚠️ IMPORTANT:
	This project is not fully completed yet, and the current weighting is not a very good tradeoff.
	If I obtain new training results in the future, I will continue to update them here

	If you have decided to test this not-so-perfect weight, please be aware：
	Because this model uses a custom architecture, you MUST pass `trust_remote_code=True` when loading it.

	### Prerequisites
	```bash
	pip install transformers torch