DESK

Loading…

Radar · Entity

DeepSeek

Everything the desk has tracked for this organisation.

Organisation

Model releases

DeepSeek-V4-Flash-0731 0731

weights unknown

DeepSeek · released Jul 31, 2026

signal

68

DeepSeek V4-Flash V4-Flash

weights unknown

DeepSeek · released Jul 31, 2026

signal

67

V4-Flash

weights unknown

DeepSeek · released Jul 31, 2026

signal

67

DeepSeek V4 Pro 0813 0813

open weights

DeepSeek

signal

86

DeepSeek-V4.1-Flash V4.1-Flash

open weights

DeepSeek

multimodal

signal

81

DeepSeek V4.1 Flash 1017 1017

open weights

DeepSeek

text

signal

63

Research papers

large language modelsexplanation reliabilitypublished Oct 26, 2026 · arXiv:2610.32665

Do Language Models Trust Their Own Justifications? A Study on Functional Consistency

reasoningintermediate-state reusepublished Oct 21, 2026 · arXiv:2610.27662

When Can LLM Reasoning Reliably Reuse Intermediate Results?

large language modelsworkflow automationpublished Oct 16, 2026 · arXiv:2610.21502

ISW-Agent: Integrating LLM-Driven Dialogue with n8n Workflow for Internal Solitary Wave Remote Sensing Image Processing

LLM agentsagent memorypublished Oct 16, 2026 · arXiv:2610.21170

MemScope: Evolving Agent Memory Architectures through Localized Code Repair

Recent source items

86 items on this desk · showing the latest 20

Do Language Models Trust Their Own Justifications? A Study on Functional ConsistencyOct 26, 2026
China clears DeepSeek, MiniMax for overseas AI deploymentOct 25, 2026
DeepSeek reportedly raises US$15B as IPO plans gather paceOct 23, 2026

reinforcement learninglarge language modelspublished Oct 9, 2026 · arXiv:2610.10942

StoreBench: A Live-Commerce Environment for Evaluating and Training Autonomous Operator Agents

agent systemsevaluation benchmarkspublished Oct 2, 2026 · arXiv:2610.00917

Finding the Right Fit: Model-Harness Interactions across Agent Tasks

Microsoft’s Mustafa Suleyman says its AI outperforms doctors on hard cases as he chases ‘medical superintelligence’Oct 22, 2026

DeepSeek reportedly raises $15B days after US restrictionsOct 22, 2026

When Can LLM Reasoning Reliably Reuse Intermediate Results?Oct 22, 2026

[AINews] DeepSeek DeepInfer: Open Source SOTA Inference running 155T toks/day/cluster making $22m/day at ~80% margin for a 81 day paybackOct 21, 2026

Latest open artifacts (#25): DeepSeek V4.1 Flash 1017, American-made models, and new regulatory controls in the US & ChinaOct 20, 2026

DeepSeek Open-Sources DeepInfer, the Inference Framework Behind DeepSeek-V4.1-FlashOct 20, 2026

US orders federal agencies to remove China’s DeepSeek AI from their systemsOct 20, 2026

😸 AI hackers left the door openOct 20, 2026

[AINews] not much happened todayOct 20, 2026

DeepSeek-Huawei chip co-design could shift AI training away from NvidiaOct 20, 2026

US orders DeepSeek removed from federal civilian IT systemsOct 20, 2026

😺 DeepSeek's free model edged Opus 5 at codingOct 19, 2026

DeepSeek updates V4.1 Flash to improve AI agent performanceOct 19, 2026

DeepSeek Just Released V4.1-Flash-1017, Scores 77.0 on DeepSWE, Opens Search APIOct 17, 2026

Anthropic opens the files on global Claude misusedate unverified