Benchmark Almanac

ByteDance Seed

33 models 15 model cards read seed.bytedance.com

What its model cards measure

Tall ticks are ByteDance Seed model cards that list the benchmark in their headline table. Gray dots are cards that leave it out. Rows are the benchmarks the lab listed most, at least twice each, in the order it started using them.

Beyond AIME GPQA Diamond SuperGPQA AIME 2025 SWE-bench Verified MMLU-Pro AIME 2024 ARC-AGI COLLIE OSWorld Online-Mind2Web AndroidWorld Windows Agent Arena ScreenSpot-Pro MATH-Vision (MathVision) MMMU-Pro TVBench LVBench Video-MME MMMU MathVista CharXiv MotionBench TOMATO OVBench OVO-Bench Humanity's Last Exam BrowseComp

Model cards

DateModelDocumentBenchmarks
23 Jun 2026 Seed2.1 Pro Seed2.1 launch page and blog (Pro and Turbo) 40
28 Apr 2026 Seed2.0 Lite (0428) Seed2.0 page, Seed2.0 Lite (0428) omni-modal upgrade tables 66
14 Feb 2026 Seed2.0 Pro Seed2.0 launch blog and model card (Pro, Lite, Mini) 94
19 Dec 2025 Seed-Prover 1.5 Seed-Prover 1.5 tech report (arXiv 2512.17260) and blog 6
18 Dec 2025 Seed1.8 Official Release of Seed1.8: A Generalized Agentic Model (blog and model card) 105
2 Sep 2025 UI-TARS-2 UI-TARS-2 Technical Report (arXiv 2509.02544) 9
20 Aug 2025 seed-oss-36b-instruct Seed-OSS-36B-Instruct model card (Hugging Face) 19
31 Jul 2025 Seed-Prover Seed-Prover tech report (arXiv 2507.23726) and GitHub README 6
31 Jul 2025 Seed Diffusion Preview Seed Diffusion Preview blog and tech report (arXiv 2508.02193) 8
25 Jun 2025 Seed1.6-Thinking Introduction to Techniques Used in Seed1.6 (blog) 11
20 May 2025 BAGEL-7B-MoT BAGEL-7B-MoT model card (Hugging Face) 8
12 May 2025 Seed1.5-VL Seed1.5-VL GitHub README and Technical Report (arXiv 2505.07062) 54
8 May 2025 Seed-Coder-8B-Instruct Seed-Coder-8B-Instruct model card (Hugging Face) 6
16 Apr 2025 UI-TARS-1.5 UI-TARS-1.5 model card (Hugging Face, UI-TARS-1.5-7B) and blog 7
10 Apr 2025 Seed-Thinking-v1.5 Seed-Thinking-v1.5 GitHub README (Full Results table) 14

Every ByteDance Seed model