Project AURIONAI Use-Cases Research & Innovation
ByteDance

ByteDance

Big tech, China. 53 models in the directory.

big techChina
0deployments

Identity

Type
big tech
Country label
China
HQ country
China
Founding country
China
Parent
not in the notes
Parent country
not in the notes
Country basis
the research notes
Units
PT TokopediaPT Tokopedia
Where this comes from
Convention D1: a dual label, HQ first, when the HQ, founding or parent country differ.

Models by Class

Models

ModelClassOpennessParametersReleased
ByteDanceSeedance 2.5video generationproprietary–31 Jul 2026ByteDanceSeedream 5.0 Proimage generationproprietary–Jul 2026ByteDanceBernini-Rvideo generationopen weights permissive–21 May 2026ByteDanceProtenix-v2scientificproprietary464M8 Apr 2026ByteDanceProtenix-v1scientificopen source368M5 Feb 2026ByteDanceSeedance 2.0video generationproprietary–Feb 2026Tsinghua UniversityTimer-S1Tsinghua University, ByteDancetime seriesunknown–2026ByteDanceSeedream 4.5image generationproprietary–Dec 2025ByteDanceSeedream 4.0image generationproprietary–Sep 2025ByteDanceOmniHuman-1.5avatarproprietary–26 Aug 2025ByteDanceProtenix-Mini / Mini+scientificopen source–16 Jul 2025ByteDanceSeed-X PPO 7Btranslationopen weights permissive7.51B16 Jul 2025ByteDanceSeed1.6multimodal LLMunknown–25 Jun 2025ByteDanceSeedVR2video generationopen weights permissive–5 Jun 2025ByteDanceSeed-Coder-8B-Instructcodeopen weights permissive8B4 Jun 2025ByteDanceSeed1.6-Thinkingreasoningproprietary–Jun 2025ByteDanceSeedance 1.0video generationproprietary–Jun 2025ByteDanceBAGEL (BAGEL-7B-MoT)multimodal LLMopen weights permissive14B20 May 2025ByteDanceBAGEL (7B MoT)image generationunknown14BMay 2025ByteDanceSeedream 3.0image generationproprietary–15 Apr 2025ByteDanceDreamActor-M1avatarproprietary–2 Apr 2025ByteDanceOmniHuman-1avatarproprietary–3 Feb 2025ByteDanceSeedVRvideo generationopen weights permissive–2 Jan 2025ByteDanceLatentSync 1.6avataropen weights restricted–Dec 2024ByteDanceProtenixscientificunknown–8 Nov 2024ByteDanceLLaVA-OneVision Qwen2 7Bmultimodal LLMopen weights permissive8.03B13 Aug 2024The University of Hong KongDepth Anything V2 SmallThe University of Hong Kong, ByteDancevisionopen weights permissive24.8M13 Jun 2024The University of Hong KongDepth Anything V2 LargeThe University of Hong Kong, ByteDancevisionopen weights restricted335.3M13 Jun 2024ByteDancePuLIDimage generationopen weights permissive–24 Apr 2024ByteDanceAnimateDiff-Lightningvideo generationopen weights restricted–19 Mar 2024ByteDanceSDXL-Lightningimage generationopen weights restricted–20 Feb 2024The University of Hong KongDepth AnythingThe University of Hong Kong, ByteDancevisionopen weights permissive–19 Jan 2024ByteDanceNext-ViTvisionunknown–12 Jul 2022Nanyang Technological UniversityoCLIPNanyang Technological University, ByteDanceOCR and documentunknown–8 Mar 2022University of WashingtonRobust Video MattingUniversity of Washington, ByteDancevisionunknown–25 Aug 2021The University of Hong KongSparse R-CNNThe University of Hong Kong, ByteDancevisionunknown–25 Nov 2020University of AdelaideSOLOUniversity of Adelaide, ByteDancevisionunknown–10 Dec 2019ByteDanceMaskTrack R-CNNByteDance, University of Illinois Urbana-Champaign, Adobe Inc.visionunknown–12 May 2019ByteDanceDola-Seed 2.0 ProLLMproprietary––ByteDanceDoubao Seed Codecodeproprietary––ByteDanceSeed-Coder-8B-Basecodeunknown8B–ByteDanceSeed2.1LLMproprietary––ByteDanceSeedancevideo generationunknown––ByteDanceSeedance 1.5 Provideo generationproprietary––ByteDanceSeedRealtimespeech synthesisunknown––ByteDanceSeedreamimage generationunknown––ByteDanceSeedream 5.0 Liteimage generationproprietary––Tsinghua Universityvideo-SALMONN 2+ 72BTsinghua University, ByteDancemultimodal LLMunknown––ByteDanceDolphin (document parsing)OCR and documentopen weights permissive––ByteDanceSeedVRvideo generationunknown––ByteDanceOmniHumanavatarunknown––ByteDanceDreamActoravatarunknown––ByteDanceDreamActor-M2.0avatarproprietary––

Deployments

1 deployments across 1 company of the group: 1 production.

Vendor countries: 1 USA. Evidence: 1 company statement.

Deployments in Indonesia

Vertex AI to improve product data qualityTokopedia, Indonesiaproduction–

Scores

BenchmarkModelScoreWho ran itSetting
Arena Text leaderboardDola-Seed 2.0 Prorank onlyindependent
Artificial Analysis Intelligence IndexDoubao Seed Code17 indexindependentasterisk on page
Artificial Analysis Text-to-Image ArenaBAGEL (BAGEL-7B-MoT)706 EloindependentRank range 161 to 163
Artificial Analysis Text-to-Image ArenaSeedream 4.01,030 Eloindependentrank shown as 20-27
Artificial Analysis Text-to-Image ArenaSeedream 4.51,023 Eloindependent
Artificial Analysis Text-to-Image ArenaSeedream 5.0 Pro1,080 Eloindependentrank shown as 17-18
Artificial Analysis Text-to-Video leaderboardSeedance 2.01,116 EloindependentDreamina Seedance 2.0
Artificial Analysis Text-to-Video leaderboardSeedance 2.51,143 EloindependentDreamina Seedance 2.5
GenEvalBAGEL (7B MoT)0.88 scorevendor-reportedpaper-reported, labelled unverified by Codesota; year 2025; with LLM rewriter
GIFT-EvalTimer-S10.485 CRPS (geo-mean, normalised to Seasonal Naive)independentpretrained; rank is average CRPS rank
GIFT-EvalTimer-S10.693 MASE (geo-mean, normalised to Seasonal Naive)independentpretrained
GIFT-EvalTimer-S126.5 mean CRPS rankvendor-reportedzero-shot foundation-model CRPS rank, lower is better
HumanEvalSeed-Coder-8B-Instruct84.8 %vendor-reported
LiveCodeBenchSeed-Coder-8B-Instruct24.7 %vendor-reportedLiveCodeBench 2410-2502
LMArena Text-to-Image ArenaSeedream 5.0 Pro1,255 Eloindependent
LMArena Text-to-Video ArenaSeedance 2.01,479 Eloindependent720p
LMArena Text-to-Video ArenaSeedance 2.51,474 Eloindependent720p
MBPPSeed-Coder-8B-Instruct85.2 %vendor-reported
MMMUBAGEL (BAGEL-7B-MoT)55.3 %vendor-reported
VBench-2.0Seedance 1.053.04 %maintainer-runCreativity score; Seedance 1.0 Pro
VBench-2.0Seedance 1.059.81 %maintainer-runTotal score; Seedance 1.0 Pro
Video-MMEvideo-SALMONN 2+ 72B81.6 %maintainer-runWith subtitles
Where this comes from
Every recorded score of this organisation's models, with who ran it. Scores from different settings are not directly comparable.

Sources

  1. ByteDance Seed seed.bytedance.com https://seed.bytedance.com/en/
  2. ByteDance-Seed/Seed-Coder-8B-Instruct , Hugging Face huggingface.co https://huggingface.co/ByteDance-Seed/Seed-Coder-8B-Instruct
  3. GitHub - bytedance/Protenix: Toward High-Accuracy Open-Source Biomolecular Structure Prediction. github.com https://github.com/bytedance/Protenix
  4. From AlphaFold3 to Protenix: Making Biomolecular Modeling More Practical neurosnap.ai https://neurosnap.ai/blog/post/from-alphafold3-to-protenix-making-biomolecular-modeling-more-practical/6772f2644fd5e47a17836a41
  5. ByteDance - Wikipedia en.wikipedia.org https://en.wikipedia.org/wiki/ByteDance
  6. Video-MME video-mme.github.io https://video-mme.github.io/home_page.html
  7. Over 30 AI models have been trained at the scale of GPT-4 epoch.ai https://epoch.ai/data-insights/models-over-1e25-flop

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "ByteDance", https://aurion.semai.id/organisations/bytedance/, records of October 2026.