Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures · HackerPapers discussions

Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures — mentioned by lhl

Sat, 09 Aug 2025 07:39:40 GMT

A few people have mentioned looking a the vLLM docs and blog (recommended!). I'd also recommend SGLang's docs and blog as well. If you're interested in a bit of a deeper dive, I can highly recommend reading some of what DeepSeek has published: https://arxiv.org/abs/2505.09343 (and actually quite a few of their Technical Reports and papers). I'd also say that while the original GPT-4 was a huge model when it was originally released (rumored 1.7T-A220B), these days you can get (original release) "GPT-4-class" performance at ~30B dense/100B sparse MoE - and almost all the leading MoEs have…

Paper and discussion history. A mention is not an endorsement.

Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures — mentioned by lexandstuff

Thu, 29 May 2025 01:16:31 GMT

Except that, all Deepseek models so far have been trained on Nvidia hardware. For Deepseek v3, they literally mention that they used 2,048 NVIDIA H800 GPUs right in the abstract: https://arxiv.org/html/2505.09343v1

Paper and discussion history. A mention is not an endorsement.

Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures — HN submission

Wed, 21 May 2025 19:19:17 GMT

Insights into DeepSeek-V3: Scaling Challenges on Hardware for AI Architectures

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures — HN submission

Sat, 17 May 2025 02:24:05 GMT

DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures

4 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures — HN submission

Thu, 15 May 2025 22:24:10 GMT

Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures — mentioned by qtwhat

Thu, 15 May 2025 11:51:00 GMT

DeepSeek-V3 demonstrates that thoughtful hardware-software co-design can overcome the scaling challenges of large language models. By integrating innovations like Multi-head Latent Attention (MLA), Mixture of Experts (MoE) architectures, FP8 mixed-precision training, and a Multi-Plane Network Topology, DeepSeek-V3 achieves cost-effective training and inference at scale. This paper delves into these advancements and discusses future directions for AI hardware and architecture co-design. Read the full paper here: https://arxiv.org/abs/2505.09343

Paper and discussion history. A mention is not an endorsement.

Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures — HN submission

Thu, 15 May 2025 11:51:00 GMT

DeepSeek-V3: Achieving Efficient LLM Scaling with 2,048 GPUs

7 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.