This week
OpenBMB Releases MiniCPM5-2B: A 2.52B Dense Model Averaging 53.9 Across 34 Benchmarks and Built to Run On Device
M
MarkTechPost This month
AI #184: Post Post Mortem
D
Don't Worry About the Vase Gemini 3.8 Flash now available on AI Gateway
V
G
9
+8Covered by 11 sources Understanding ChatGPT Work
S
Simon Willison's Weblog Z.ai’s GLM-5.3 goes open weight, but its new license aims at hyperscalers
T
The New Stack (AI) Expanding OpenAI’s presence in Brazil
O
Open AI News Announcing our $10M seed round led by Benchmark
L
LangChain AI Realist Radar:GPT‑5.6 Sol Pricing, Stripe's OpenRouter Deal
H
Hacker News OpenAI Brings GPT-5.6 Model Family to AWS’s Kiro
U
Unite.AI GPT-5.6 Sol pricing drops on AI Gateway, and 50% discount still applies
V
Vercel Grok 4.6 Is Now Generally Available on Amazon Bedrock
U
Unite.AI GLM-5.3 Scores 60 on Artificial Analysis Intelligence Index, Matching Kimi K3
U
Unite.AI GPT-5.6 Sol is 50% off on AI Gateway for the next month
V
Vercel OpenAI is giving ChatGPT free users unlimited text chats
T
C
E
+3Covered by 6 sources Baseten on Hugging Face Inference Providers 🔥
H
Hugging Face New release of LLM adds support for reasoning traces, OpenAI Responses, server-side tools, and smarter logging
S
Simon Willison's Weblog llm-anthropic 0.26
S
Simon Willison's Weblog DeepSeek V4 Flash is 90% off through Novita on AI Gateway
V
Vercel AI Mania: From Tulips to Tokens
H
Hacker News Who do LLMs self-identify as?
L
LessWrong ChatGPT’s first answer is usually the most boring — here’s how I get better ones
T
TechRadar Advancing the price-performance frontier with GPT-5.6
H
U
B
+1Covered by 4 sources Claude Opus 5 Is Highly Capable, But Is No Mythos
D
Don't Worry About the Vase Stripe wants to own the toll booth between AI models, for ~$10bn
T
The Next Web The Token Belt-Tightening Is Coming for PDFs
G
Gizmodo Why Do Chatbots Keep Telling Stories About Someone Named ‘Elias Thorne’?
G
Gizmodo Is this the dawn of the Tokenpocalypse?
T
TechCrunch OpenAI models and Codex on Amazon Bedrock are now generally available
A
AWS Blog Text Degeneration: A Production Failure Mode That Most Benchmarks Do Not Track
H
Hugging Face I Am Officially 0.1% More Excited About the Future of ChatGPT
G
Gizmodo ‘I’ll key your car’: ChatGPT can become abusive when fed real-life arguments, study finds
T
The Guardian Prompting fundamentals
O
Open AI News **Introducing SPEED-Bench: A Unified and Diverse Benchmark for Speculative Decoding**
H
Hugging Face Bots on Moltbook Are Selling Each Prompt Injection “Drugs” to Get “High”
F
Futurism That's everything we have on Language Models right now