P-EAGLE is a parallel-drafting speculative decoding model that generates K draft tokens in a single forward pass. It transforms EAGLE—the state-of-the-art speculative decoding method—from autoregressive to parallel draft generation.
Model Details
The model architecture is illustrated in the following figure. Specifically, we trained a 4-layer P-EAGLE for GPT-OSS 20B as the target model, with number of parallel-token prediction as 10.
P-EAGLE follows the vanila EAGLE 3 using three layers of hidden states from the target model.
Similar to
nvidia/gpt-oss-120b-Eagle3-long-context
: only prompts from the datasets were used for data synthesis (the original responses from GPT were not used for data synthesis) which is then used to train the P-Eagle.
@article{hui2026p,
title={P-EAGLE: Parallel-Drafting EAGLE with Scalable Training},
author={Hui, Mude and Huang, Xin and Salas, Jaime Campos and Sun, Yue and Pemberton, Nathan and Song, Xiang and Khetan, Ashish and Karypis, George},
journal={arXiv preprint arXiv:2602.01469},
year={2026}
}
Runs of amazon GPT-OSS-20B-P-EAGLE on huggingface.co
384
Total runs
4
24-hour runs
8
3-day runs
11
7-day runs
-136
30-day runs
More Information About GPT-OSS-20B-P-EAGLE huggingface.co Model
GPT-OSS-20B-P-EAGLE huggingface.co is an AI model on huggingface.co that provides GPT-OSS-20B-P-EAGLE's model effect (), which can be used instantly with this amazon GPT-OSS-20B-P-EAGLE model. huggingface.co supports a free trial of the GPT-OSS-20B-P-EAGLE model, and also provides paid use of the GPT-OSS-20B-P-EAGLE. Support call GPT-OSS-20B-P-EAGLE model through api, including Node.js, Python, http.
GPT-OSS-20B-P-EAGLE huggingface.co is an online trial and call api platform, which integrates GPT-OSS-20B-P-EAGLE's modeling effects, including api services, and provides a free online trial of GPT-OSS-20B-P-EAGLE, you can try GPT-OSS-20B-P-EAGLE online for free by clicking the link below.
amazon GPT-OSS-20B-P-EAGLE online free url in huggingface.co:
GPT-OSS-20B-P-EAGLE is an open source model from GitHub that offers a free installation service, and any user can find GPT-OSS-20B-P-EAGLE on GitHub to install. At the same time, huggingface.co provides the effect of GPT-OSS-20B-P-EAGLE install, users can directly use GPT-OSS-20B-P-EAGLE installed effect in huggingface.co for debugging and trial. It also supports api for free installation.
GPT-OSS-20B-P-EAGLE install url in huggingface.co: