Every week, I organize 400 to 700 links into roughly 60 categories as part of my ongoing effort to learn about AI. This is my personal notebook, which I enjoy sharing with friends… a hobby and a labor of love, rather than a commercial publication or product.
If you arrived here through a search or shared link, this page collects the links I found for DeepSeek for the week ending August 28, 2026.
As part of my learning process, I like to automate the category covers. It gives me a chance to learn Python and APIs.
This week’s cover prompt was written using Claude Opus 4.7, and the image was generated using Nano Banana 2.1.
Category Cover Image Prompt: Mid-1980s murder-mystery comedy movie poster, 16:9, pure black background, a broad three-story Victorian mansion with twin turrets and gabled center rendered in high-contrast black-and-white pen-and-ink engraving with dense crosshatching; warm amber windows show costumed suspects in theatrical poses (a sly researcher in a mandarin-collar tuxedo holding an origami whale, a quant in velvet smoking jacket with slide rule and teacup, a hooded figure in dark silk clutching a labeled hard drive) plus windows crowded with stacked research papers tied in red string, abacuses beside GPU chips, a chessboard, and a paper whale lantern; center double doors open on warm light with the same slim silver-haired butler in black tailcoat holding a silver tray with an envelope sealed by a tiny wax whale; fat rounded extra-heavy white sans-serif title 'DeepSeek' above the roofline, campy spooky theatrical mood.
This Week in DeepSeek News
Summary by Claude Sonnet 5.5 based on this week’s links:
- A vision model for V4-Flash: DeepSeek put an experimental multimodal model, DeepSeek-V4-Flash-Vision-Exp, on its API. The company says it matches V4-Flash on text, and The Decoder reports it rivals Opus 4.8 on agent benchmarks. Images are billed at up to 384 tokens each at V4-Flash pricing.
- Developer plumbing: A free Files API lets you upload an image once and reuse it by file_id, which saves request bandwidth. The vision model works with the Chat Completions, Messages and Responses formats.
- DeepSeek as a design reference: Posts this week treat DeepSeek's architecture as a benchmark for other labs. Elie Bakouch notes other Chinese frontier models have moved to linear attention while DeepSeek has not. Sebastian Raschka says GLM-5.3-Flash (the model behind the Ox Alpha name) uses DeepSeek Sparse Attention layers. On price, one post puts DeepSeek-V4 Flash 0731 at 53% for $0.46 per task, against GLM-5.3-Flash at 63% for $0.24.
This summary was generated by Claude Sonnet 5.5 to help you explore the links below. Rest assured, I select, organize, and check the links by hand in Google Sheets, and write the introduction and personal commentary in The Main Newsletters myself each week as a labor of love.
This week's links related to DeepSeek
Deepseek releases experimental Flash vision model that rivals Opus 4.8 on agent benchmarks
https://the-decoder.com/deepseek-releases-experimental-flash-vision-model-that-rivals-opus-4-8-on-agent-benchmarks/
every chinese frontier model now uses linear attention (except deepseek) they all use (except kimi) sparse attention with similar indexer/compression designs to maximize efficiency they all use “fancy” residuals (mHC, attention residual, gated residual) to maximize signal”
https://x.com/eliebakouch/status/2092622716046107132
Now we know: The popular Ox Alpha LLM was GLM-5.3-Flash… Compared to GLM-5.2, this new GLM-5.3-Flash model uses: – a Kimi Linear-style 3:1 (super*) hybrid attention pattern with 34 Kimi Delta Attention layers (KDA) and 11 Multi-heat Latent Attention (MLA) / DeepSeek Sparse”
https://x.com/rasbt/status/2092629415813365899
DeepSeek-V4-Flash-Vision-Exp is now live on the DeepSeek API Platform! 🚀 🔹 This experimental multimodal model matches DeepSeek-V4-Flash on text capabilities… including agents, reasoning, and world knowledge. 🔹 On multimodal agent benchmarks, V4-Flash-Vision-Exp makes a major”
https://x.com/deepseek_ai/status/2090730032574631962
2026 Chinese model landscape in one list ▪️ Qwen3.8-27B, Qwen3.8-Max ▪️ GLM-5.3 ▪️ Kimi K3 ▪️ DeepSeek-V4 ▪️ ERNIE 5.1 ▪️ Baichuan-M3-235B, Baichuan-Omni-1.5 ▪️ Yi 1.5 ▪️ MiniCPM-o 4.5, MiniCPM-V 4.5, MiniCPM-V 4.6 They’re getting interesting in very different ways. Here we”
https://x.com/TheTuringPost/status/2091665921849098363
GLM-5.3-Flash aka Ox Alpha is now the best intelligence per dollar choice now! > GLM-5.3-Flash – 63% for $0.24/task > DeepSeek-V4 Flash 0731 – 53% for $0.46/task > GPT-5.6 Luna – 67% for $0.60/task”
https://x.com/zainhas/status/2092709719966400694
China’s Hackers Use DeepSeek for Attacks, Researchers Say – Bloomberg
https://www.bloomberg.com/news/articles/2026-08-24/chinese-hackers-use-deepseek-to-boost-attacks-researchers-say-mt7o4205
Multimodal API support 🔌 🔹 Set model=’deepseek-v4-flash-vision-exp’ 🔹 Images are tokenized for billing: up to 384 tokens each, at V4-Flash pricing 🔹 Supports Chat Completions, Messages & Responses 🔹 Supports mixed text + image input; images can be provided via base64,”
https://x.com/deepseek_ai/status/2090730039973392531
Files API is now live. 📁 🔹 Free to use 🔹 Upload an image once, then reference it by file_id to save request bandwidth 🔹 Reuse the same image across requests… no need to upload it again Learn more:
https://t.co/68OKIq6xmc 4/n”
https://x.com/deepseek_ai/status/2090730042586489333
[2501.12948] DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
https://arxiv.org/abs/2501.12948
Leave a Reply