DeepSeek V4 Flash

Über dieses Modell

DeepSeek V4 Flash is a 284 B parameter Mixture-of-Experts (MoE) chat model from DeepSeek AI. It features a hybrid attention architecture with compressed sparse and heavily compressed attention, supporting a 1 million token context window. The model achieves 88.7 % EM on MMLU, 69.5 % Pass@1 on HumanEval, and 44.7 % EM on LongBench‑V2, demonstrating strong language, coding, and long‑context capabilities. It is released under the MIT License.

Technische Daten

Fähigkeiten

Eingabe-Modalitäten

Ausgabe-Modalitäten

Reasoning

Hybrid Standard on

Knowledge horizon