The story so far
Weights, every dated article, and repo commits on one rail.
22 repo commits since the weights landed
Teal is the weights (2022-03-02), grey ticks are repo commits.
bart-large
facebook Apache 2.0 out 5 years ago
Text in, embeddings out, 1K tokens of context. 1 provider serves it.
Verified 2026-10-25 Rank #54 in Embeddings · score 88.7 method v1 Sources 3 Downloads 30d 138.1k · #282 of 619 tracked
Do people use it?138.1kdownloads, 30 days#282 of 619 we track · top 46%+1 likes, downloads −8% this week207 likesHugging Face2026-10-24
Where to run it
Live hosted endpoints, cheapest first, USD per million tokens. Ties break on output price, then provider name.
| # | Cheapest live row | In /1M | Out /1M | Context | Quant | Seen |
|---|---|---|---|---|---|---|
| 1 | HF Inference | — | — | — | — | Oct 25 |
Where it's used
Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.
Served by 1 API provider; 1 quantized build for local use; a big base for builders (235 fine-tunes, adapters and merges).
- As a service1HF Inference
- On own hardware1quantized builds
- Built on235203 fine-tunes · 31 adapters · 1 merges
- In apps100+Hugging Face Spaces
Top Spaces running it: ByteDance/XVerse ♥ 25 · enclap-team/enclap ♥ 13 · Thouph/Furrence-2-Large-Demo ♥ 11 · HaloMaster/chinesesummary ♥ 9
What people built on it
236 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.
236 derivatives on the Hub, most of them fine-tunes
From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.
Reference
Everything else, folded: the full tables, every source.
What the card says
2 cited sentences
Summary generated, cited
facebook/bart-large is a transformer encoder-decoder model pre-trained on the English language.[1][2] The architecture consists of a bidirectional encoder and an autoregressive decoder.[3]
- model type
- “BART model pre-trained on English language.”
- “bidirectional (BERT-like) encoder and an autoregressive (…”
Generated 2026-10-07 15:04 UTC by parasail-gemma-4-26b-a4b-it; each numeral points at the fact it rests on. Not a verdict.
Use-case evidence
1 use case
What it is good at 1 use case
Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.
Facts
17 fields
Listed in Embeddings
Facts
- Context length is 1,024 tokens (1K). (config.json max_position_embeddings, observed 2026-10-24)
- License: Apache 2.0 (friction: none). Permissive; patent grant. (license file, observed 2026-10-24)
- Created 2022-03-02, last modified 2022-06-03. (HF API, observed 2026-10-24)
| Pipeline tag | feature-extraction |
|---|---|
| Library | transformers |
| Architectures | BartModel |
| Model type | bart |
| MoE | no |
| Context length | 1,024 (config.json max_position_embeddings) |
| Languages | en |
| Papers | arXiv:1910.13461 |
| License | Apache 2.0 (apache-2.0) |
| Hub snapshot | 2026-10-25 |
Servability
vLLM · SGLang
Servability
vLLM: not in registry · BartModel · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: not in registry · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed
vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.
All variants
40 variants
Lineage
finetune: 203adapter: 31quantized: 1merge: 1
Top variants (40 shown, ranked by downloads)
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| ilsilfverskiold/tech-keywords-extractor | finetune | — | 406M | S | apache-2.0 | ilsilfverskiold | 38 | 237.5k |
| Nevidu/LexBartLo_1 | adapter | — | — | — | — | Nevidu | 0 | 26.1k |
| ilsilfverskiold/bart-keyword-extractor | finetune | — | 406M | S | apache-2.0 | ilsilfverskiold | 7 | 8.5k |
| whyredfire/legal-bart-summarizer | finetune | — | 406M | S | apache-2.0 | whyredfire | 0 | 6.1k |
| uzw/bart-large-question-generation | finetune | — | 406M | S | apache-2.0 | uzw | 1 | 4.9k |
| cive202/humanize-ai-text-bart-large | finetune | — | 406M | S | mit | cive202 | 3 | 3.5k |
| Intel/bart-large-mrpc | finetune | — | — | — | apache-2.0 | Intel | 0 | 1.5k |
| pszemraj/bart-large-summary-map-reduce | finetune | — | 406M | S | apache-2.0 | pszemraj | 2 | 1.1k |
| marcelomoreno26/bart-large-samsum-adapter | adapter | — | — | — | apache-2.0 | marcelomoreno26 | 0 | 887 |
| eilamc14/bart-large-text-simplification | finetune | — | 406M | S | apache-2.0 | eilamc14 | 0 | 886 |
| Short-Answer-Feedback/bart-finetuned-saf-communication-networks | finetune | — | 406M | S | — | Short-Answer-Feedback | 2 | 768 |
| tobijen/bart_left_heading_torch | finetune | — | — | — | apache-2.0 | tobijen | 0 | 817 |
| shahin-as/bart-large-sentence-compression | finetune | — | 406M | S | — | shahin-as | 1 | 703 |
| taiypeo/bart-large-wikilarge | finetune | — | 406M | S | apache-2.0 | taiypeo | 0 | 635 |
| xshubhamx/bart-large | finetune | — | 407M | S | apache-2.0 | xshubhamx | 0 | 616 |
| dmis-lab/ANGEL_pretrained | finetune | — | — | — | gpl-3.0 | dmis-lab | 5 | 361 |
| taiypeo/bart-large-wiki-doc-full | finetune | — | 406M | S | apache-2.0 | taiypeo | 1 | 524 |
| elzinga/stories_model_bert_v2 | finetune | — | 406M | S | apache-2.0 | elzinga | 0 | 569 |
| taiypeo/bart-large-d-wikipedia | finetune | — | 406M | S | apache-2.0 | taiypeo | 0 | 566 |
| bpavlsh/bart-crypto-summary | finetune | — | 406M | S | — | bpavlsh | 0 | 544 |
| tobijen/bart_left_fulltext_torch | finetune | — | — | — | apache-2.0 | tobijen | 0 | 529 |
| taiypeo/bart-large-d-wikipedia-full | finetune | — | 406M | S | apache-2.0 | taiypeo | 1 | 456 |
| taiypeo/bart-large-wiki-doc | finetune | — | 406M | S | apache-2.0 | taiypeo | 1 | 412 |
| mongrz/cspl_translation | finetune | — | 406M | S | apache-2.0 | mongrz | 0 | 454 |
| Lynote/cive202-humanize-ai-text-bart-large-browser | quantized | — | — | — | mit | Lynote | 0 | 452 |
| serkanarslan/text2text-example | finetune | — | — | — | apache-2.0 | serkanarslan | 0 | 420 |
| bhargavis/fulltrain-xsum-bart | finetune | — | 406M | S | mit | bhargavis | 1 | 346 |
| SrujanTopalle/ViT-Bart_Image_Captioner | finetune | — | 341M | S | apache-2.0 | SrujanTopalle | 1 | 335 |
| adamjweintraut/bart-finetuned-loaf-512-lyrictoplan | finetune | — | 406M | S | apache-2.0 | adamjweintraut | 0 | 379 |
| genejalston/bart-qanda2 | finetune | — | 406M | S | apache-2.0 | genejalston | 0 | 353 |
| Supreeth40/finetuned-bartL-samsum | finetune | — | 406M | S | apache-2.0 | Supreeth40 | 0 | 344 |
| adamjweintraut/bart-finetuned-loaf-256-lyrictoplan | finetune | — | 406M | S | apache-2.0 | adamjweintraut | 1 | 285 |
| zwellington/bart-pubhealth-expanded | finetune | — | — | — | apache-2.0 | zwellington | 0 | 331 |
| adamjweintraut/bart-finetuned-eli5_lfqa_top | finetune | — | 406M | S | apache-2.0 | adamjweintraut | 0 | 323 |
| rudra496/stealthhumanizer-bart | finetune | — | 406M | S | mit | rudra496 | 0 | 323 |
| Koshti10/BART-large-ET-Synthetic | finetune | — | — | — | apache-2.0 | Koshti10 | 0 | 312 |
| learn3r/bart_large_gov | finetune | — | 406M | S | apache-2.0 | learn3r | 0 | 304 |
| aiguy68/legal_text_summarizer | finetune | — | 60.5M | S | apache-2.0 | aiguy68 | 0 | 302 |
| varun-v-rao/bart-large-squad-model2 | finetune | — | 406M | S | apache-2.0 | varun-v-rao | 0 | 297 |
| yunu919/bart-large-dialogue-summarization | finetune | — | 406M | S | mit | yunu919 | 1 | 230 |
Repository changelog
22 commits
Repository changelog 22 commits
| Date | Commit | SHA |
|---|---|---|
| 2022-06-03 | typo: encoder-encoder -> encoder-decoder (#1) | cb48c13 |
| 2022-03-09 | Update config.json | 030bb1b |
| 2021-09-16 | Add model card | 9e16983 |
| 2021-06-14 | add flax model | 76041a4 |
| 2021-03-09 | Move tokenizer.json from roberta-large | 22fa338 |
| 2020-12-11 | Migrate model card from transformers-repo | b51e601 |
| 2020-10-26 | Update config.json | f8275ef |
| 2020-10-15 | Update tf_model.h5 | 6000698 |
| 2020-09-10 | Update pytorch_model.bin | b47d1df |
| 2020-08-25 | Update tokenizer_config.json | f15be28 |
| 2020-08-25 | Update vocab.json | 78e492d |
| 2020-08-25 | Update merges.txt | 2ee0f9b |
| 2020-04-25 | Update rust_model.ot | 6cc4a1f |
| 2020-04-24 | Update config.json | 5c11bea |
| 2020-04-02 | Update config.json | e052e84 |
| 2020-03-30 | Update config.json | 379b637 |
| 2020-03-14 | Update config.json | 3c69355 |
| 2020-02-13 | Update config.json | 6968574 |
| 2020-02-13 | Update config.json | 2073576 |
| 2020-02-13 | Update config.json | e090573 |
| 2020-02-12 | Update config.json | aa29da2 |
| 2020-02-12 | initial commit | b543227 |
All providers
1 provider · 1 endpoint
Where to run 1 providers · 1 endpoints
Hosted endpoints observed for bart-large, sorted by input price (USD per million tokens). Each row names its observer and date; prices change daily, so check the source before quoting.
| Provider | Model ref | Context | Quant | $/M in | $/M out | tok/s | TTFT | Tools | Status | Observed |
|---|---|---|---|---|---|---|---|---|---|---|
| HF Inference hf | facebook/bart-large | — | — | — | — | — | — | — | live | 2026-10-25 |
Papers
1 paper
Papers 1
arXiv links found in the model card and its tags. "Peer-reviewed" is shown only with evidence: a published venue in OpenAlex (measured) or the authors' own journal reference on arXiv (reported).
- BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension this model peer-reviewed · measuredMike Lewis, Yinhan Liu, Naman Goyal +5 · 2019-10-29 · Annual Meeting of the Association for Computational Linguistics (ACL) · PDF · HF paper page (6 upvotes) · evidence
Similar models
3 models
Similar models
- bge-large-zh-v1.5 same task (feature-extraction), permissive license
- codebert-base same task (feature-extraction)
- wavlm-large same task (feature-extraction)
Sources
3 sources
Sources
Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:36 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:45 UTC · Summary: 2026-10-07 15:04 UTC · Last verified 2026-10-25
Follow bart-large
Watch this model
- /feeds/models/facebook--bart-large.xml(RSS 2.0) · every change to bart-large
- /feeds/families/facebook--bart-large.xml(RSS 2.0) · every change across the bart-large family
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.