DeepSeek V4 Flash.

A 284B agentic engine for the rare Mac that can host a private, large-scale thinking partner—five shards of weight for a workspace that refuses to leave the workstation.

View on Hugging Face See LocalEngine variants ↓

Model makerDeepSeek

Supported in LocalEnginemacOS

Upstream licenseMIT

SourceDeepSeek V4 Flash model card ↗

Where it earns its space

Built for a real local rhythm.

These are good-fit starting points—not promises of benchmark performance. Choose the smallest listed option that comfortably covers your work.

Complex private agent workflows

Large-scale code and reasoning tasks

Air-gapped high-memory Mac setups

In LocalEngine

Configured variants, by device.

The formats below are the downloads currently represented in LocalEngine’s model catalogs. GGUF runs through the llama.cpp-based runtime; MLX options use the Apple Silicon MLX runtime.

macOS

284B

Formats: GGUF UD-Q4_K_XL · UD-Q8_K_XL
Local note: Q4 is about 155 GB and calls for 192 GB unified memory; Q8 is about 162 GB and calls for 208 GB.

About the source

Model identity, task category, supported formats, and license are based on the linked Hugging Face DeepSeek V4 Flash model card and the LocalEngine macOS/iOS catalogs. Hardware guidance reflects the packaged variants, not a universal system requirement.