LMSYS Org Blog company blog
Deploying DeepSeek on 96 H100 GPUs
DeepSeek is a popular open-source large language model (LLM) praised for its strong performance. However, its large size and unique architecture, which uses Multi-head Latent Attention (MLA) and Mixture of Experts (MoE), require an advanced system for efficient serving at scale.
- Points
- 285
- HN comments
- 80
- Submissions
- 1
- Cited by
- 3 commenters
AI & Machine LearningInfrastructure & DevOps2025
On Hacker News
| Submitted | Title on HN | Points | Comments |
|---|---|---|---|
| Aug 29, 2025 | Deploying DeepSeek on 96 H100 GPUs by GabrielBianconi | 285 | 80 |
Cited in comments
Linked from 5 Hacker News comments by 3 different people, from Jun 2025 to Nov 2025.
Most recent:
