Model obtained by Fine Tuning 'facebook/bart-large-xsum' using AMI Meeting Corpus, SAMSUM Dataset, DIALOGSUM Dataset, XSUM Dataset!
BART model pre-trained on English language, and fine-tuned on CNN Daily Mail. It was introduced in the paper BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension by Lewis et al.
Parameters406M
Context1,024
Weights8.5 GB
Licensemit
AccessOpen weights
Monthly Downloads1.3M
Runs On
What it takes to serve bart-large-cnn (406M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 0.8 GB | 1.0 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.4 GB | 0.5 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.2 GB | 0.2 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.
SAVRN's Notes on bart-large-cnn
For condensing English news-style text, a 406M-parameter summarizer trained on cnn_dailymail is the right size. At 16-bit it wants 0.8 GB of weights and 1.0 GB of memory. The cheapest listed setup is one MI300X with 192 GB at $1.85 an hour on-demand, far more card than one copy needs, so we would batch it beside heavier work. Watch the download: 8.54 GB across 13 files, because safetensors, pytorch, jax, rust and tf all ship together; pull only the one your runtime loads.
MIT is as light as a license gets: commercial use, modification and redistribution, provided the notices ride along. The real constraint is the 1,024-token context, so anything longer than an article gets chunked first. Check two things: the reported ROUGE-1 of 42.9 is on cnn_dailymail 3.0.0, its own training set, and the model card was not written by the publisher's team.
Model Card
By AI at Meta, published under mit, revision 37f520fa929c.
BART (large-sized model), fine-tuned on CNN Daily Mail
BART model pre-trained on English language, and fine-tuned on CNN Daily Mail. It was introduced in the paper BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension by Lewis et al. and first released in [this repository (https://github.com/pytorch/fairseq/tree/master/examples/bart).
Disclaimer: The team releasing BART did not write a model card for this model so this model card has been written by the Hugging Face team.
Model description
BART is a transformer encoder-encoder (seq2seq) model with a bidirectional (BERT-like) encoder and an autoregressive (GPT-like) decoder. BART is pre-trained by (1) corrupting text with an arbitrary noising function, and (2) learning a model to reconstruct the original text.
BART is particularly effective when fine-tuned for text generation (e.g. summarization, translation) but also works well for comprehension tasks (e.g. text classification, question answering). This particular checkpoint has been fine-tuned on CNN Daily Mail, a large collection of text-summary pairs.
Intended uses & limitations
Configuration
- Architecture
- BartForConditionalGeneration
- Context length (tokens)
- 1,024
- Layers
- 12
- Vocabulary size
- 50,264
- Model type
- bart
Identity and Version
- Repository
- facebook/bart-large-cnn
- Publisher
- AI at Meta
- Task
- Summarization
- Modality
- Text
- Library
- transformers
- Parameters
- 406M parameters
- Languages
- en
- Revision
- 37f520fa929c961707657b28798b30c003dd100b
- First published
- 2022-03-02
- Last updated
- 2024-02-13
Files and Weights
13 files, 8.5 GB in total. The weights are 5 files totalling 8.5 GB in bin, h5, msgpack, ot, safetensors.
Weights5 files · 8.5 GB
Configuration3 files · 2.3 KB
Tokenizer3 files · 2.7 MB
Documentation1 file · 6.0 KB
Repository1 file · 445 B
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| flax_model.msgpack | Weights | 1.6 GB | dda1068890b2 |
| model.safetensors | Weights | 1.6 GB | 40041830399a |
| pytorch_model.bin | Weights | 1.6 GB | 2ac2745c02ac |
| rust_model.ot | Weights | 2.0 GB | cd0d1586babf |
| tf_model.h5 | Weights | 1.6 GB | 8d845bfe2bbb |
| config.json | Configuration | 1.6 KB | — |
| generation_config.json | Configuration | 363 B | — |
| generation_config_for_summarization.json | Configuration | 363 B | — |
| README.md | Documentation | 6.0 KB | — |
| .gitattributes | Repository | 445 B | — |
| merges.txt | Tokenizer | 456.3 KB | — |
| tokenizer.json | Tokenizer | 1.4 MB | — |
| vocab.json | Tokenizer | 898.8 KB | — |
License and Download
- License
- mit
- Access
- Open weights, no gate
- Download size
- 8.5 GB
Download from AI at Meta
Released by AI at Meta through its official repository on Hugging Face. Read the license.
Built From
- Described by arXiv:1910.13461
- Trained on (disclosed) cnn_dailymail
Evaluations
Each result is shown as reported, with the conditions its reporter stated. None is a SAVRN measurement. A comparison lines two results up only when their configuration, unit and setup are all stated and identical.
| Benchmark | Conditions | Result | Reported by | Revision | Date |
|---|---|---|---|---|---|
| cnn_dailymail | Configuration 3.0.0Task SummarizationMetric ROUGE-1Comparison conditions not established | 42.9486 | facebook Publisher reported |
Evaluated revision not stated | — |
| cnn_dailymail | Configuration 3.0.0Task SummarizationMetric ROUGE-2Comparison conditions not established | 20.8149 | facebook Publisher reported |
Evaluated revision not stated | — |
| cnn_dailymail | Configuration 3.0.0Task SummarizationMetric ROUGE-LComparison conditions not established | 30.6186 | facebook Publisher reported |
Evaluated revision not stated | — |
| cnn_dailymail | Configuration 3.0.0Task SummarizationMetric ROUGE-LSUMComparison conditions not established | 40.0376 | facebook Publisher reported |
Evaluated revision not stated | — |
| cnn_dailymail | Configuration 3.0.0Task SummarizationMetric gen_lenComparison conditions not established | 78.5866 | facebook Publisher reported |
Evaluated revision not stated | — |
| cnn_dailymail | Configuration 3.0.0Task SummarizationMetric lossComparison conditions not established | 2.529 | facebook Publisher reported |
Evaluated revision not stated | — |
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 8.5 GB |
| 16-bit | 0.8 GB |
| 8-bit | 0.4 GB |
| 4-bit | 0.2 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About bart-large-cnn
How much GPU memory does bart-large-cnn need?
About 1 GB at 16-bit and 0.2 GB at 4-bit: the weights (406M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run bart-large-cnn on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use bart-large-cnn commercially?
Yes. bart-large-cnn is released under MIT License. The MIT License is a short permissive license. It permits commercial use, modification and redistribution, provided the copyright notice and permission notice are included.
What is bart-large-cnn's context length?
1,024 tokens, from the maximum position embeddings in its published configuration.
Similar Models
This model is a fine-tuned version of allenai/led-large-16384 on the BookSum dataset (kmfoda/booksum). It aims to generalize well and be useful in summarizing lengthy text for both academic and everyday purposes. - See the Colab demo linked above or try the demo on Spaces To improve summary quality, use encodernorepeatngramsize=3 when calling the pipeline object. This setting encourages the model to utilize new vocabulary and construct an abstractive summary. Load the model into a pipeline object: Feed the text into the pipeline object: Important: For optimal summary quality, use the global attention mask when decoding, as demonstrated in this community notebook, see the definition of…
Imagine you have a very long news article, and you want someone to read it and tell you the key points in just a few sentences. That's exactly what this model does! This model takes a long news article and turns it into a short, easy-to-read summary. - It writes back a 3-4 sentence summary that captures all the important information The special thing about this model is that it's: 1. Very accurate - It understands news writing style very well 2. Very fast - It works quickly even on regular computers (not just expensive AI servers) 3. Specialized in news - It was trained specifically on news articles, so it understands how journalists write 4. Good with financial news - It knows market…
This model was fine-tuned on a novel financial news dataset, which consists of 2K articles from Bloomberg, on topics such as stock, markets, currencies, rate and cryptocurrencies. It is based on the PEGASUS model and in particular PEGASUS fine-tuned on the Extreme Summarization (XSum) dataset: google/pegasus-xsum model. PEGASUS was originally proposed by Jingqing Zhang, Yao Zhao, Mohammad Saleh and Peter J. Liu in PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization. Note: This model serves as a base version. For an even more advanced model with significantly enhanced performance, please check out our advanced version on Rapid API. The advanced model offers more…
Finetuned ai-forever/ruT5-base for text and dialogue summarization. All 'train' subsets was concatenated and shuffled with seed 1000 - 7. Train subset = 155678 rows. Evaluation on 10% of concatenated 'validation' subsets = 1458 rows. See WandB logs. See report at REPORT WIP.
This model is a fine-tuned version of Google's T5-Base model tailored for summarizing financial 10K report sections. T5-Base-10K-Summarization is optimized to condense lengthy 10K reports into manageable summaries, enabling quick insights into financial data and trends. Ideal for use by financial analysts and regulatory agencies needing rapid insights from 10K reports. It may not be suited for summarizing non-financial documents or informal texts. Trained on a diverse collection of 10K reports from various industries, annotated for summarization to ensure broad applicability and accuracy. The following hyperparameters were used during training: - learningrate: 5e-05 - trainbatchsize: 8…