DeepSeek R1 0528
Released 2025-05 · reasoning · 128K tokens · 7 benchmarks · Open weight
Editorial notes
Major upgrade vs R1: AIME 2024 79.8 -> 91.4, AIME 2025 70 -> 87.5, GPQA 71.5 -> 81.
Spec sheet
- Company
- DeepSeek
- Country
- CN
- Type
- reasoning
- Release
- 2025-05
- Context
- 128K tokens
- Params total
- 685B
- Params active (MoE)
- 37B
- License
- deepseek-license
- Quants
- BF16, Q8_0, Q5_K_M, Q4_K_M
- Pricing (deepseek)
- $0.55/$2.19/M
- Slug
- deepseek-r1-0528
Benchmarks (7)
Reasoning 2
Coding 2
Cite this model
BibTeX · APA
BibTeX
@misc{frontier-deepseek-r1-0528,
title = {DeepSeek R1 0528},
author = {{DeepSeek}},
year = {2025},
note = {Frontier Benchmarks AI atlas. Accessed 2026-05-08},
url = {https://frontierbenchmarks.com/models/deepseek-r1-0528}
} APA
DeepSeek (2025). DeepSeek R1 0528 [Large language model]. Frontier Benchmarks AI. Retrieved 2026-05-08, from https://frontierbenchmarks.com/models/deepseek-r1-0528
Citation reflects the atlas page, not the original model paper. For the paper, see the "Resources" section above.