Text Generation
PEFT
Safetensors
English
code
compression
coding-agent
lora
sft
claude-code
cursor
conversational
Eval Results (legacy)
Instructions to use paritok/paritok-4b-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use paritok/paritok-4b-v1 with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-4B-Instruct-2507") model = PeftModel.from_pretrained(base_model, "paritok/paritok-4b-v1") - Notebooks
- Google Colab
- Kaggle
docs: give the release a proper @software entry, credited
Browse filesThe card pointed at a bare repository URL for the weights, which credits nobody and cannot be pasted into a bibliography. A @software entry does not need a DOI — author, title, version and url are enough — and it matches the entry in the GitHub README and CITATION.cff.
README.md
CHANGED
|
@@ -204,7 +204,18 @@ Cite the **paper** for the method and results:
|
|
| 204 |
}
|
| 205 |
```
|
| 206 |
|
| 207 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 208 |
|
| 209 |
---
|
| 210 |
|
|
|
|
| 204 |
}
|
| 205 |
```
|
| 206 |
|
| 207 |
+
Cite the **software** for this release — the weights, the gateway, and the training and evaluation code:
|
| 208 |
+
|
| 209 |
+
```bibtex
|
| 210 |
+
@software{paritok-4b-v1,
|
| 211 |
+
title = {Paritok-4B-v1: An Open-Source Compression Model for AI Coding Agents},
|
| 212 |
+
author = {Shi, Jiayu and Chen, Luzhuo},
|
| 213 |
+
year = {2026},
|
| 214 |
+
version = {4B-v1},
|
| 215 |
+
license = {Apache-2.0},
|
| 216 |
+
url = {https://github.com/Paritok-official/paritok-4b-v1},
|
| 217 |
+
}
|
| 218 |
+
```
|
| 219 |
|
| 220 |
---
|
| 221 |
|