DeepSeek-V4: a million-token context that agents can actually use

Hugging Face Blog · 2026-04-24

Chinese Original

Related items

AIHugging Face Blog2026-03-09

Ulysses Sequence Parallelism: Training with Million-Token Contexts

AIHugging Face Blog2025-01-28

Open-R1: a fully open reproduction of DeepSeek-R1

AIHugging Face Blog2024-12-31

Introducing smolagents: simple agents that write actions in code.

AIHugging Face Blog2023-01-24

What Makes a Dialog Agent Useful?

AIHugging Face Blog2025-01-31

Mini-R1: Reproduce Deepseek R1 „aha moment“ a RL tutorial

AIHugging Face Blog2025-04-04

Journey to 1 Million Gradio Users!

Feedback

TypeMessage