DeepSeek V4 Flash

deepseek-ai · released 2026-04-24 · mit license

DeepSeek-V4-Flash is the lighter 284B (13B active) member of the V4 family with 1M context (April 2026, MIT). In max-reasoning mode it approaches the Pro on coding, lagging on knowledge and the hardest agentic work.

Key specs

TypeLocal open-weight
Parameters158.07B total · MoE, — active
Architecturedeepseek_v4
Context window1049K tokens
Knowledge cutoff
Modalitiestext
Recommended backends
Minimum viable rig

Benchmark scores

GPQA Diamond88.1%
SWE-bench Verified79%
AIME
MMLU-Pro86.4%
BFCL v3 (tool use)
Composite score7.09
Community ratingNo reviews yet

VRAM & disk per quantization

QuantVRAMDiskRAMContext
Q4_K_M93.2 GB91.7 GB1049K

Strengths & weaknesses

Strengths: 284B total / 13B active MoE with 1M context, very efficient; Strong coding for its active size (LiveCodeBench 91.6); MIT licensed

Weaknesses: Knowledge and hardest agentic tasks trail the Pro variant; Scores depend heavily on reasoning-effort mode; vendor-reported only