Model

DeepSeek-V4.1-Flash

OPEN

DeepSeek · DeepSeek-V4 family · released Sep 10, 2026

DeepSeek's 2026-09-10 release, the smallest model of a new architecture family: a 552B-backbone causal encoder-decoder MoE that activates ~8B parameters in prefill and 16B in decode, with native image input, a 1M context and a KV cache about a quarter the size of V4-Flash's. Open weights under MIT. On DeepSeek's own table it beats every compared model on Terminal-Bench 2.1 (90.6), DeepSWE v1.1 (74.2), Codeforces (3471), CyberGym (88.1) and AutomationBench (54.8), but trails the closed frontier on HLE and Terminal-Bench 3.0/4.0. The API alias deepseek-flash now points to it, legacy V4-Flash names are routed to it, and from 2026-09-14 deepseek-v4-pro requests are also served by V4.1-Flash at Flash rates until V4.1-Pro launches.

ReasoningCodingVisionFunction callingTool useAgentic

2754.1

Elo · rank #20

Parameters

552B

Active params

16B (MoE)

Context

1M tokens

Architecture

552B sparse MoE, 40-layer causal encoder-decoder (1 shared + 384 routed experts, 6 routed active per token), ~8B active in prefill / 16B in decode, ~890 bytes/token global KV cache (about 1/4 of V4-Flash); natively multimodal (image+text in, text out); ~45T multimodal pre-training tokens

License

MIT

Languages

—

API price (in/out)

$0.3 / $1.2

Modalities

text · vision

Benchmark results

Bar shows position within the tracked field; marker = field best

AutomationBenchAgents54.8%#1

best: this model · 54.8%