DeepSeek-V3 Technical Report (arXiv 2412.19437)

Fields

id

sources/sched-deepseek-v3-report

URL

https://arxiv.org/abs/2412.19437

Publisher

arXiv

Kind

paper

Published

2024-12

Accessed

2026-10-04

Archived URL

https://arxiv.org/html/2412.19437v2

notes

DeepSeek-V3 technical report. Relevant for expert parallelism (EP) vs tensor parallelism (TP) discussion: DeepSeek-V3 uses 256 experts with top-8 routing, and the report discusses EP as the preferred parallelism strategy for MoE models. The arXiv abstract was fetched and verified.

Notes

DeepSeek-V3 technical report. Relevant for expert parallelism (EP) vs tensor parallelism (TP) discussion: DeepSeek-V3 uses 256 experts with top-8 routing, and the report discusses EP as the preferred parallelism strategy for MoE models. The arXiv abstract was fetched and verified.

Cross-references

This record references no other records.