Latest
GPT-6 Reaches Every ChatGPT Tier: Luna for Free Users, Sol for Paid
OpenAI is rolling GPT-6 into ChatGPT's Chat tab for every plan: GPT-6 Sol for paid tiers from 7 October, GPT-6 Luna for Free and Go from 8 October, plus "Intelligent UI" answers with charts, forms and mini tools. What changed, which Sol this is, and what OpenAI's own safety card flags.
NVIDIA’s LoGRA Cuts RL Training Memory With Low-Rank Gradient Sketches
NVIDIA’s LoGRA paper shows low-rank gradient sketches cutting average RL training memory by up to 45.7% on tested Qwen reasoning setups—and making 27B RL runnable on one 8×H100 node where dense Adam OOMs. What the method is, what Table 1 shows, and what it does not prove.
Mistral Large 4 Preview: Artificial Analysis scores it 38
Third-party firm Artificial Analysis scored Mistral Large 4 Preview 38 on Intelligence Index v4.3.2. Company training and open-weight claims are labelled separately.
Grok Bot’s Best-Backend Shift: Outcomes Over In-House-Only
Elon Musk says Grok Bot will use the best back-end model for each task—naming Claude Opus 5.5, MidJourney, Suno and other leading APIs. What he stated, what remains unconfirmed, and why routing details still matter.
Mistral Large 4 and Qwen3.8 Show Open-Source AI Now Ships in Months
Open weights, serving runtimes and agents now fit together. Why open-source AI adoption is compressing from years of glue work to months.
South Korea’s bank hacks: possible AI use raises the stakes
South Korea’s president ordered a full response after customer data was exposed at seven financial firms, saying some cases showed possible AI use. Police have formed a 28-member investigation team.
Reflection’s Beam: an open-weight workhorse pitched on efficiency
Reflection AI introduced Beam, a sparse open-weight Mixture-of-Experts model with 501 billion total parameters and 23 billion active, aimed at coding and agentic work. Weights are promised later this month; scores are company-reported.
Gemini 4 Argon: Google’s Benchmark Claims Meet Insider Coding Doubts
Google says Gemini 4 Argon is its next frontier model, with record benchmark scores and a 1M-token output limit. Most people can't use it yet, and some Google staff, speaking anonymously, question its coding. Here's what's claimed and what's checked.
OpenAI’s GPT-6.1 Sol Nears GPT-6 Astra at a Fifth of the Price
OpenAI says GPT-6.1 Sol nearly matches GPT-6 Astra at about a fifth of the price. Here's what is on the official price list, what OpenAI's safety paperwork shows, and what is still only a claim.
Another Open Model, Not a New Leader: What Reflection’s Reported Release Would Change
If the Axios scoop is right, the open-weight field gets another entrant, and the top of the US market does not. Reflection has not confirmed it, and there is still nothing to download.
Latest on TSN
Explore the TSN Atlas: Every topic we cover, mapped. Click a bubble to dive in.
