ChunkLlama2-13B

Closed weights Alibaba,The University of Hong Kong,Fudan University 13B parameters May 2024

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
Alibaba,The University of Hong Kong,Fudan University
Organisation type
Industry,Academia,Academia
Country
China, Hong Kong
Published
29 May 2024
Authors
Chenxin An, Fei Huang, Jun Zhang, Shansan Gong, Xipeng Qiu, Chang Zhou, Lingpeng Kong

What it does

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

Domain
Language
Task
Language modeling/generation, Question answering
Base model
Llama 2-13B

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
13B

same as base model

Training data
256,000,000 tokens

The training dataset is sourced from ShareGPT and AlpacaGPT (Taori et al., 2023). For the data derived from ShareGPT, we specifically curate a subset by extracting responses generated by GPT-4, and dialogues that exceed 4k tokens in length. This selection results in a compilation of 5,405 training instances. We further finetune Llama2 with over 16k steps with a batch size of 1. prompt length 16k tokens ->16000 steps*16000 tokens= 256000000 tokens

Training compute

The arithmetic performed to train the model, measured in floating-point operations. It is a measure of what the training run cost, not of how fast the finished model answers you.

How it was established
Hardware,Operation counting
Fine-tuning compute
2 × 10¹⁹ FLOP

6 FLOP/parameter/token * 13000000000 parameters * 256000000 tokens = 19968000000000000000 FLOP 312000000000000 FLOP/GPU/sec * 60 GPU-hours * 3600 sec / hour * 0.3 [assumed utilization] = 20217600000000000000 FLOP sqrt(19968000000000000000*20217600000000000000) = 2.0092412e+19

The training run

What it physically took to train: which chips, how many, for how long, and what that drew from the wall.

Training hardware
NVIDIA A100 SXM4 80 GB
Chip-hours
60

Availability

Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.

Weights
Closed — provider access only
Model access
Unreleased
Training code
Open source

Apache 2.0 https://github.com/HKUNLP/ChunkLlama

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
Training-Free Long-Context Scaling of Large Language Models
Last updated
28 November 2025

What the numbers mean

Background

ChunkLlama2-13B was published by Alibaba,The University of Hong Kong,Fudan University, in the country recorded as China, during May 2024. The category the publisher falls under is industry,Academia,Academia.

It works in the domain of Language, and is recorded as performing the task of language modeling/generation, Question answering.

It builds on Llama 2-13B. That is the usual way a specialised model is produced.

Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.

Training and provenance

Training consumed a corpus of around 256,000,000 tokens of text.

Answers

ChunkLlama2-13B — common questions

01

ChunkLlama2-13B— what GPU do I need to run it?

None. This 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.

02

ChunkLlama2-13B— is it open source?

No. Its weights have not been published, so it exists only as a service controlled by its owner.

03

ChunkLlama2-13B— how many parameters does it have?

It has a parameter count of 13B. same as base model. 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.

04

ChunkLlama2-13B— who created it?

It was published by Alibaba,The University of Hong Kong,Fudan University, based in China, an organisation categorised as industry,Academia,Academia.

05

ChunkLlama2-13B— when was it released?

It was published in May 2024. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

06

ChunkLlama2-13B— what is it used for?

It works in the domain of Language, and is recorded as handling the task of language modeling/generation, Question answering. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

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.