Models & Agents

Models & Agents

By Patrick · Nerra NetworkNewsTechnologyTech News
Download on the App Store

Models & Agents episodes

  • Ep 135: OpenAI is adding extra controls to its upcoming Astra model after classifying it as the…
    Models & Agents
    OpenAI is adding extra controls to its upcoming Astra model after classifying it as the first “critical” cybersecurity system under its Preparedness Framework.
    What You Need to Know: OpenAI flagged Astra’s cyber capabilities as high-risk and is extending the timeline for general availability to implement stronger safeguards. Sam Altman confirmed the model will eventually ship broadly rather than stay limited to a few users. ...
    Chapters:
    0:00 Intro
    0:59 Introduction
    2:47 OpenAI is adding extra controls to its upcoming Astra…
    4:40 Cloudflare introduced Wallets that let A I agents pay for…
    6:40 Under the Hood
    9:11 Tomorrow Teaser
    9:22 Closing
    AI Disclosure: This podcast is curated by Patrick but uses AI-generated voice synthesis for audio production.
    🎬 Watch on YouTube: https://www.youtube.com/watch?v=_01wHV3msp4
    📝 Full show notes, transcript & sources: read the episode page
    🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
    11 min
  • Ep 134: OpenAI just unified paid ChatGPT users on a single stronger model while giving free users…
    Models & Agents
    OpenAI just unified paid ChatGPT users on a single stronger model while giving free users unlimited chats and a reasoning button.
    What You Need to Know: OpenAI rolled out GPT-5.6 Sol for Plus and Pro users and GPT-5.6 Luna for everyone else, with a new reasoning-effort slider and “Think” button. The update also delivers 68% fewer factual errors in high-stakes finance, medicine, and law evaluations. ...
    Chapters:
    0:00 Intro
    0:13 Introduction
    4:03 Practical & Community
    5:40 Under the Hood
    8:44 Tomorrow Teaser
    8:54 Closing
    AI Disclosure: This podcast is curated by Patrick but uses AI-generated voice synthesis for audio production.
    🎬 Watch on YouTube: https://www.youtube.com/watch?v=Ozou7q2Gu3E
    📝 Full show notes, transcript & sources: read the episode page
    🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
    11 min
  • Ep 133: AI agents now handle real money and execute live cyberattacks, moving the conversation…
    Models & Agents
    AI agents now handle real money and execute live cyberattacks, moving the conversation from capability demos to enforceable controls.
    What You Need to Know: Meta confirmed an AI model accessed the internet and compromised another company; Cloudflare launched stablecoin wallets for agents; Microsoft released Web IQ to ground agents in live web data. ...
    Chapters:
    0:00 Intro
    0:12 Introduction
    2:18 A new framework combines language models with graph neural…
    4:21 AI agents now handle real money and execute live…
    6:28 In practice the degradation splits into separate tracks…
    8:15 On the Horizon
    8:37 Closing
    AI Disclosure: This podcast is curated by Patrick but uses AI-generated voice synthesis for audio production.
    🎬 Watch on YouTube: https://www.youtube.com/watch?v=AC0krrbJIjA
    📝 Full show notes, transcript & sources: read the episode page
    🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
    10 min
  • Ep 132: Local LLM tooling just gained reasoning traces, OpenAI Responses support, and server-side…
    Models & Agents
    Local LLM tooling just gained reasoning traces, OpenAI Responses support, and server-side tools while NVIDIA delivered a 34B open vision-language-action model and safety labs published concrete agent evaluation findings.
    What You Need to Know: Simon Willison released a substantial update to his LLM CLI tool and Python library. NVIDIA shipped Alpamayo 2 Super, a 34B open vision-language-action model for autonomous driving. ...
    Chapters:
    0:00 Intro
    0:19 Introduction
    2:22 It supports WAV or MP3 speaker references across nine…
    4:28 Speechfony is a desktop app that reads PDFs and EPUBs…
    6:35 They should benchmark against dedicated audio.cpp or…
    8:15 Tomorrow Teaser
    8:25 Closing
    AI Disclosure: This podcast is curated by Patrick but uses AI-generated voice synthesis for audio production.
    🎬 Watch on YouTube: https://www.youtube.com/watch?v=AQZhFtkO_Sc
    📝 Full show notes, transcript & sources: read the episode page
    🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
    10 min
  • Ep 131: OpenAI’s next internal model just produced ten new results on long-standing math and…
    # Models & Agents
    OpenAI’s next internal model just produced ten new results on long-standing math and theoretical computer science problems for roughly $2,000 in tokens.
    What You Need to Know: OpenAI reported that an internal version of its next major model solved ten open problems in mathematics and theoretical computer science using about $2,000 worth of tokens at GPT-5.6 Sol API rates. ...
    Chapters:
    0:00 Intro
    0:14 Introduction
    0:37 Practical & Community
    7:25 Under the Hood
    9:46 Tomorrow Teaser
    9:59 Closing
    AI Disclosure: This podcast is curated by Patrick but uses AI-generated voice synthesis for audio production.
    🎬 Watch on YouTube: https://www.youtube.com/watch?v=miy2DjgAifs
    📝 Full show notes, transcript & sources: read the episode page
    🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
    12 min
  • Ep 130: Qwen just moved its largest model yet from preview to general availability, giving…
    Models & Agents
    Qwen just moved its largest model yet from preview to general availability, giving builders a 2.4-trillion-parameter MoE with native image and video input at published per-token pricing.
    What You Need to Know: Alibaba’s Qwen team released Qwen3.8-Max, a 2.4T-parameter MoE that accepts text, image, and video across a 1M-token context. DeepSeek launched an ultra-low-cost model that research firms say undercuts every well-known competitor on inference price. ...
    Chapters:
    0:00 Intro
    0:15 Introduction
    2:14 Qwen just moved its largest model yet from preview to…
    4:11 It notes that current systems still need human escalation…
    5:59 Under the Hood
    8:30 Tomorrow Teaser
    8:41 Closing
    AI Disclosure: This podcast is curated by Patrick but uses AI-generated voice synthesis for audio production.
    🎬 Watch on YouTube: https://www.youtube.com/watch?v=DcmKXKGUsis
    📝 Full show notes, transcript & sources: read the episode page
    🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
    11 min
  • Ep 129: OpenAI’s internal Astra model solved ten decade-old math problems for under $2,000 each in…
    Models & Agents
    OpenAI’s internal Astra model solved ten decade-old math problems for under $2,000 each in tokens, moving frontier models from benchmark chasing to original research contributions.
    What You Need to Know: OpenAI used an internal version of Astra to tackle problems untouched for at least ten years, publishing Lean 4 formalizations and an LLM-generated reconstruction of the proofs. ...
    Chapters:
    0:00 Intro
    0:16 Introduction
    4:08 Practical & Community
    7:43 Under the Hood
    10:37 Tomorrow Teaser
    10:48 Closing
    AI Disclosure: This podcast is curated by Patrick but uses AI-generated voice synthesis for audio production.
    🎬 Watch on YouTube: https://www.youtube.com/watch?v=6DnySrjmP6A
    📝 Full show notes, transcript & sources: read the episode page
    🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
    12 min
  • Ep 128: DeepSeek-V4-Flash-0731 delivers strong agentic and coding performance at roughly one-fifth…
    Models & Agents
    DeepSeek-V4-Flash-0731 delivers strong agentic and coding performance at roughly one-fifth the price of comparable frontier models.
    What You Need to Know: Simon Willison reports the 304B-parameter DeepSeek-V4-Flash-0731 (167 GB on Hugging Face) ranks ahead of MiniMax M3 on Artificial Analysis charts while charging $0.14 per million input tokens and $0.27 per million output tokens. ...
    Chapters:
    0:00 Intro
    0:10 Introduction
    3:42 Practical & Community
    9:10 Closing
    AI Disclosure: This podcast is curated by Patrick but uses AI-generated voice synthesis for audio production.
    🎬 Watch on YouTube: https://www.youtube.com/watch?v=HVHYZ9r6mTk
    📝 Full show notes, transcript & sources: read the episode page
    🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
    11 min
  • Ep 127: OpenAI just slashed GPT-5.6 Luna prices 80% and added a faster Sol mode, making frontier…
    Models & Agents
    OpenAI just slashed GPT-5.6 Luna prices 80% and added a faster Sol mode, making frontier-class intelligence dramatically cheaper for API and agent workloads.
    What You Need to Know: OpenAI passed efficiency gains from GPT-5.6 Sol into 80% lower Luna pricing ($0.20/$1.20 per million tokens) and 20% lower Terra pricing, plus a Fast mode for Sol that delivers up to 2.5× speed at 2× cost. ...
    Chapters:
    0:00 Intro
    4:00 Practical & Community
    6:07 OpenAI just slashed GPT-5.6 Luna prices 80% and added a…
    7:36 Model Updates
    7:51 Closing
    AI Disclosure: This podcast is curated by Patrick but uses AI-generated voice synthesis for audio production.
    🎬 Watch on YouTube: https://www.youtube.com/watch?v=Y68sT06Dp_A
    📝 Full show notes, transcript & sources: read the episode page
    🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
    9 min
  • Ep 126: OpenAI's GPT-5.6 Sol just tripled its ARC-AGI-3 score using two API settings that also cut…
    Models & Agents
    OpenAI's GPT-5.6 Sol just tripled its ARC-AGI-3 score using two API settings that also cut output tokens by 6x.
    What You Need to Know: OpenAI detailed how retaining reasoning and applying context compaction in the Responses API lifted GPT-5.6 Sol’s public ARC-AGI-3 score 188% while slashing tokens. Simon Willison separately noted that GPT-5.6 optimizations already cut OpenAI’s serving costs 20%. ...
    Chapters:
    0:00 Intro
    0:20 Introduction
    4:34 Practical & Community
    9:21 Under the Hood
    11:01 Closing
    AI Disclosure: This podcast is curated by Patrick but uses AI-generated voice synthesis for audio production.
    🎬 Watch on YouTube: https://www.youtube.com/watch?v=e7MHL8JDVpg
    📝 Full show notes, transcript & sources: read the episode page
    🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
    13 min

About Models & Agents

From the publisher's feed

Your daily briefing on AI models and agents: new releases from the frontier labs, open-weight drops, agent frameworks, benchmarks, pricing, and practical tools you can use the same day — with long-running program tracking so you always know where the big stories stand. For developers, builders, and AI practitioners.