Eagle 2.5

Closed weights NVIDIA,Nanjing University,Hong Kong Polytechnic University,Rutgers University 8B parameters April 2025

No estimate

No hardware requirements for this model

The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.

On record

Full specification

Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.

Origin

Who built this model, where, and when it was published.

Organisation
NVIDIA,Nanjing University,Hong Kong Polytechnic University,Rutgers University
Organisation type
Industry,Academia,Academia,Academia
Country
United States of America, China, Hong Kong
Published
21 April 2025
Authors
Guo Chen, Zhiqi Li, Shihao Wang, Jindong Jiang, Yicheng Liu, Lidong Lu, De-An Huang, Wonmin Byeon, Matthieu Le, Tuomas Rintamaki, Tyler Poon, Max Ehrlich, Tuomas Rintamaki, Tyler Poon, Tong Lu, Limin Wang, Bryan Catanzaro, Jan Kautz, Andrew Tao, Zhiding Yu, Guilin Liu

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Vision, Robotics, Language
Base model
Qwen2.5-7B

Size

How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.

Parameters
8B

Table 8. "Full Model, 8B" Could be Qwen2.5-7B + MLP Connector (40M) https://nvlabs.github.io/EAGLE/ as of 2025-05-29: "Eagle-2.5 Weights (Coming Soon)"

Training data
tokens

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Record confidence
Confident

Sources

Where this record came from and when it was last checked.

Reference
Eagle 2.5: Boosting Long-Context Post-Training for Frontier Vision-Language Models
Last updated
28 November 2025

What the numbers mean

Background

Eagle 2.5 was published by NVIDIA,Nanjing University,Hong Kong Polytechnic University,Rutgers University, in United States of America, in April 2025. The organisation is categorised as industry,Academia,Academia,Academia.

It works in Vision, Robotics, Language.

Its starting point was Qwen2.5-7B — most models at this scale are adapted from an existing base rather than built from nothing.

Its weights were never published, so it can only be reached through its provider. No graphics card changes that.

Answers

Eagle 2.5 — common questions

01

When was Eagle 2.5 released?

Eagle 2.5 was published in April 2025.

02

What is Eagle 2.5 used for?

Eagle 2.5 works in Vision, Robotics, Language. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

03

What GPU do I need to run Eagle 2.5?

None. Eagle 2.5 is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.

04

Is Eagle 2.5 open source?

The licensing for Eagle 2.5 was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.

05

How many parameters does Eagle 2.5 have?

Eagle 2.5 has 8B parameters. Table 8. "Full Model, 8B" Could be Qwen2.5-7B + MLP Connector (40M) https://nvlabs.github.io/EAGLE/ as of 2025-05-29: "Eagle-2.5 Weights (Coming Soon)". That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.

06

Who created Eagle 2.5?

Eagle 2.5 was published by NVIDIA,Nanjing University,Hong Kong Polytechnic University,Rutgers University, based in United States of America, categorised as industry,Academia,Academia,Academia.

Source

Original publication

Record last updated 28 November 2025

The other direction

Looking at it from the other side?

This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.