---
type: "raw_signal"
id: "the-technical-report-describes-a-40-layer-causal-5ef881"
slug: "the-technical-report-describes-a-40-layer-causal-5ef881"
title: "The technical report describes a 40-layer Causal Encoder-Decoder network divided evenly between encoder and decoder layers."
description: "DeepSeek-AI’s technical report introduces DeepSeek-V4.1-Flash, a multimodal Mixture-of-Experts model with 552 billion backbone parameters and context support up to one million tokens."
retrieval_nugget: "DeepSeek-AI’s technical report introduces DeepSeek-V4.1-Flash, a multimodal Mixture-of-Experts model with 552 billion backbone parameters and context support up to one million tokens."
observed_at: "2026-10-17"
record_date: "2026-10-17"
date_kind: "observed_at"
why_it_matters: "Source: https://arxiv.org/pdf/2609.19969"
novelty: "notable"
verification_level: "source-linked"
signal_type: "research"
source_platform: "arxiv.org"
topics: []
entities: ["DeepSeek-AI","DeepSeek-V4.1-Flash","Compressed Sparse Attention 2","DeepSWE v1.1","Automation-Bench"]
related_patterns: ["agent-economics-moves-to-completed-work"]
source_url: "https://arxiv.org/pdf/2609.19969"
source_urls: ["https://arxiv.org/pdf/2609.19969"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "signal:the-technical-report-describes-a-40-layer-causal-5ef881"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/signals/the-technical-report-describes-a-40-layer-causal-5ef881/","markdown":"https://newruntime.com/signals/the-technical-report-describes-a-40-layer-causal-5ef881.md","json":"https://newruntime.com/signals/the-technical-report-describes-a-40-layer-causal-5ef881.json"}
---

# The technical report describes a 40-layer Causal Encoder-Decoder network divided evenly between encoder and decoder layers.

## Retrieval answer

DeepSeek-AI’s technical report introduces DeepSeek-V4.1-Flash, a multimodal Mixture-of-Experts model with 552 billion backbone parameters and context support up to one million tokens.

DeepSeek-AI’s technical report introduces DeepSeek-V4.1-Flash, a multimodal Mixture-of-Experts model with 552 billion backbone parameters and context support up to one million tokens. The report describes a 40-layer Causal Encoder-Decoder architecture, split evenly between encoder and decoder layers. It says the model activates 16 billion parameters per token during decode and 8 billion during prefill.

The design targets KV-cache storage and bandwidth costs in long-horizon workloads. DeepSeek combines cross-layer KV reuse in Compressed Sparse Attention 2 with FP4 KV caching, reporting a global KV-cache footprint of 890 bytes per token, approximately one quarter of DeepSeek-V4-Flash’s footprint. The report also says SWA Bounded Replay reduces persistent KV-cache footprint to roughly one eighth of the previous model’s. DeepSeek reports evaluation scores including 74.2 on DeepSWE v1.1 and 54.8 on Automation-Bench. These are claims and results in the authors’ report; the supplied source does not establish independent replication or production outcomes.

Source: https://arxiv.org/pdf/2609.19969

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Raw Signals](https://newruntime.com/signals.json): Open every source-linked raw signal.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)