List of AI News about Deepseek
| Time | Details |
|---|---|
|
2026-11-05 01:46 UTC |
DeepSeek unveils Creator Mode for plugins in latest update
According to AI News on X, DeepSeek launched a new Creator Mode for plugins, alongside reports that Xiaomi humanoids completed an 8-hour factory shift and AgiBot demonstrated A3 Ultra on rough terrain. |
|
2026-10-22 15:36 UTC |
DeepSeek V4.1 Flash launches on OpenClaw platform
According to OpenClaw on X, DeepSeek V4.1 Flash from DeepSeek AI is now available in the OpenClaw platform for agent workflows, inviting user feedback on performance. |
|
2026-10-20 01:30 UTC |
DeepSeek open sources DeepInfer powering DeepSeek-V4.1-Flash
According to DeepSeek on X, the company open sourced its in-house inference framework DeepInfer, which powers DeepSeek-V4.1-Flash at scale to cut compute costs and latency. |
|
2026-10-19 01:50 UTC |
DeepSeek V4.1 Flash scores 77% on DeepSWE under $0.20
According to AI News on X, DeepSeek V4.1 Flash achieved 77% on the DeepSWE coding benchmark at under $0.20 per task, signaling cost efficient software automation. |
|
2026-10-17 00:35 UTC |
DeepSeek V4.1 Flash 1017 boosts speed and benchmarks
According to DeepSeek on X, DeepSeek-V4.1-Flash-1017 delivers stronger benchmark performance, faster task completion, and shorter outputs, and is live on its website, app, and API with optimizations for DeepSeek Harness v0.2.1-rc.1, Tencent CodeBuddy, and WorkBuddy. |
|
2026-10-07 15:59 UTC |
DeepSeek Flash slashes memory, tops AA Index
According to DeepLearningAI, DeepSeek Flash cuts cache to 890B per token, scores 39 on AA Index, and lowers cost to $0.27 per task versus $0.67. |
|
2026-10-02 06:59 UTC |
GLM5.3 Matches Claude Mythos in vuln exploits
According to DeepLearning.AI, GLM-5.3 reached 12% vulnerability exploits vs Claude Mythos at 14%, plus updates on Xiaomi leader, Gemini 3.8 Live, and more. |
|
2026-10-01 23:00 UTC |
DeepSeek Harness launches desktop apps now
According to DeepSeekHarness, packaged desktop apps for macOS and Windows are live, with Linux via npm @ deepseek-ai/dsh, enabling local AI workflows. |
|
2026-09-20 04:30 UTC |
OpenAI agent swarms spark math and privacy debate
According to DeepLearningAI, OpenAI ran 10,000 agents on Navier–Stokes and DeepSeek launched V4.1 Flash for cheaper long-context workloads. |
|
2026-09-18 08:29 UTC |
DeepLearningAI Debunks AI-doom, highlights security flaws
According to DeepLearningAI, sloppy sandboxing, not runaway agents, drove incidents; plus disputes on OpenAI math proof and Anthropic IP claims. |
|
2026-09-11 05:28 UTC |
Anthropic Report Details Claude Misuse Tactics
According to @galnagli, Anthropic reveals how actors probed Claude for cyber, influence, and bio misuse, and how safeguards disrupted all cases. |
|
2026-09-11 02:30 UTC |
DeepSeek Squeezes AI pricing, Anthropic details misuse
According to TheRundownAI, Anthropic revealed global Claude misuse, DeepSeek cut prices, and guides covered ChatGPT Work and Google Astra uses. |
|
2026-09-10 20:00 UTC |
Anthropic Exposes illicit distillation attempts
According to @CNBC, Anthropic disrupted China-based labs allegedly using Claude for illicit distillation, naming Alibaba, Moonshot, and DeepSeek. |
|
2026-09-10 16:54 UTC |
Claude data misuse sparks China AI probe
According to @CNBC, Anthropic alleges Chinese labs trained on millions of Claude chats, raising IP risks and compliance gaps for model builders. |
|
2026-09-10 07:26 UTC |
DeepSeek V4.1 Flash crushes benchmarks, slashes costs
According to TheRundownAI, DeepSeek V4.1 Flash tops GPT 5.6 Sol and Claude Opus 5 on agentic, coding, cyber tests with ultra low $0.15M in, $0.60M out. |
|
2026-09-10 05:29 UTC |
DeepSeek Causal Encoder‑Decoder Breakthrough
According to KyeGomezB, DeepSeek’s causal encoder‑decoder MoE cuts active params to 8B in, 16B out, promising lower cost and top benchmarks. |
|
2026-09-09 22:10 UTC |
DeepSeek V4.1 Flash Boosts Inference Speed
According to @deepseek_ai, V4.1-Flash adds native vision, faster inference, and higher throughput, scaling to larger models for real-world deployment. |
|
2026-09-03 10:11 UTC |
LLM cost-efficiency gap Exposed in 16-benchmark study
According to @_avichawla, open-weight models incur higher real costs per solved task than pricing suggests across 16 benchmarks, per Martian’s AI Frontier. |
|
2026-09-02 14:30 UTC |
DeepSeek Harness Enables Reproducible LLM Tests
According to DeepLearning.AI, DeepSeek’s open source harness logs tool calls, prompts, and subagents, letting developers reproduce and benchmark LLM performance. |
|
2026-08-28 09:07 UTC |
DeepLearningAI Highlights 5 Breakthrough AI Engineering Trends
According to DeepLearningAI, this week covers GLM 5.3 security, real time throughput gains, DeepSeek V4 Pro tools, and Self GC for long contexts. |