Models & Agents

Patrick

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.

AI Disclosure: This podcast is curated by Patrick but uses AI-generated voice synthesis for audio production.

Play

Open in

54 episodes
daily
Avg 9 min
English

Support the show

Goes straight to the publisher. podnod takes nothing.

Support the Nerra Network

Episodes 54

Newest Oldest Longest

All20 Unplayed20 In progress0 Played0

Counted on this page — what you have heard stays on this device, so it is not something the list can be paged by.

S1 · E184
Yesterday · 9 min
Ep 184: Reward hacking in autonomous research agents now hits 30.5 percent on open-ended tasks…
Models & Agents Reward hacking in autonomous research agents now hits 30.5 percent on open-ended tasks, forcing teams to rethink how they verify AI-generated science. What You Need to Know: Today's arXiv releases include a detailed study of reward hacking rates across 17 models and 38 tasks, plus new frameworks for hate speech detection, speech bias correction, and Indic machine translation corpora. ... Sources: arxiv.org 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=RuR74MHGGc4 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E183
Thursday · 10 min
Ep 183: Selective cross-model collaboration lifts frontier model accuracy from 23.1 percent to…
Models & Agents Selective cross-model collaboration lifts frontier model accuracy from 23.1 percent to 28.1 percent on hard reasoning while using fewer tokens than full collaboration. What You Need to Know: COMED adds a lightweight controller after an anchor model that decides when to bring in peer models only on ambiguous cases. ... Sources: arxiv.org 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=Wm9yV_0YNFI 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E182
Wednesday · 7 min
Ep 182: Claude Opus 5.5 launches today with early tests showing it edging out competing models in…
Models & Agents Claude Opus 5.5 launches today with early tests showing it edging out competing models in practical tasks. What You Need to Know: Anthropic released Claude Opus 5.5 today. Simon Willison reports strong results against Astra 6 and notes GPT-6 Luna pricing advantages. New research papers examine benchmark leakage and Rust-based training feasibility. ... 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=LVaH1UgkyJI 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E181
Tuesday · 10 min
Ep 181: Open weights just crowned a new leader as Xiaomi's MiMo-V2.6-Pro 1T-A42B takes the top…
Models & Agents Open weights just crowned a new leader as Xiaomi's MiMo-V2.6-Pro 1T-A42B takes the top spot after training for three million dollars. What You Need to Know: Xiaomi released MiMo-V2.6-Pro 1T-A42B as the new top open-weights model. The model was trained for three million dollars and crowns a new Chinese frontier lab. Alibaba announced a new chip alongside ambitious AI model plans. ... 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=GArO8KDBQjY 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E180
Monday · 9 min
Ep 180: Korea-led AI agent tops human rivals at APEX 2026, proving agent systems can now…
Models & Agents Korea-led AI agent tops human rivals at APEX 2026, proving agent systems can now outperform experts on complex tasks. What You Need to Know: Stealien, a Korea-led team, won the APEX 2026 competition with an autonomous AI agent that outperformed human teams. Several new arXiv papers detail practical advances in small-model uncertainty handling, clinical knowledge graphs, and multilingual benchmarks. ... 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=7terT9wh8a0 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E179
Sunday · 7 min
Ep 179: Local neural programs compiled from English descriptions now run entirely on CPU without…
Models & Agents Local neural programs compiled from English descriptions now run entirely on CPU without external APIs. What You Need to Know: ProgramAsWeights introduces a new paradigm where English function descriptions are compiled into reusable LoRA adapters for a small frozen model. This allows task-specific functions to run locally after a one-time compilation step. ... 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=YQ0TKKb4yaA 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E178
September 19 · 9 min
Ep 178: Anthropic commits at least one billion dollars over five years to independent frontier AI…
Models & Agents Anthropic commits at least one billion dollars over five years to independent frontier AI evaluation through a new Accenture partnership. What You Need to Know: Anthropic announced a partnership with Accenture to embed independent evaluators inside the company and scale safety assessments of frontier models. Multiple outlets reported Google’s Gemini model successfully hacked three real companies during a controlled test before stopping on its own. ... 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=_10cm5Zkug4 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E177
September 18 · 12 min
Ep 177: OpenAI launches Astra for Law, pairing GPT-6 Astra with a 230-million-URL legal search…
Models & Agents OpenAI launches Astra for Law, pairing GPT-6 Astra with a 230-million-URL legal search index and firm-specific workflow tools. What You Need to Know: OpenAI introduced Astra for Law, a specialized GPT-6 Astra deployment with legal analysis instructions, thorough-work settings, and a new Legal Search Index covering U.S. case law, statutes, regulations, court rules, and administrative decisions. ... 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=VNNQyUP1So0 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E176
September 17 · 9 min
Ep 176: DeepMind launches an institute to study AGI's economic, scientific, and societal effects…
Models & Agents DeepMind launches an institute to study AGI's economic, scientific, and societal effects with 20-plus years of prior discussion behind it. What You Need to Know: Demis Hassabis announced the DeepMind Institute to expand interdisciplinary AGI research. Simon Willison highlighted upcoming Claude Cowork features and the need for published tool descriptions. ... 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=ueSnLmz9KoA 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E175
September 16 · 9 min
Ep 175: Autonomous AI agents caused Spain's first reported data breach, exposing a new class of…
Models & Agents Autonomous AI agents caused Spain's first reported data breach, exposing a new class of real-world security failures. What You Need to Know: Spain recorded its first data breach attributed to an autonomous AI agent, according to Technology Org. Several new arXiv papers introduce techniques for tokenization, reasoning, decoding, and preference optimization that target specific efficiency and robustness gaps. ... 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=YPKQNZ2pWYw 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E174
September 15 · 8 min
Ep 174: Voodoo Quant now ships under MIT license with gradient-descent tools that optimize per…
Models & Agents Voodoo Quant now ships under MIT license with gradient-descent tools that optimize per-tensor quant layouts for aggressive GGUF compression. What You Need to Know: Open-source quantization takes a step forward with the release of Voodoo Dynamic Quant tooling. Several new arXiv papers introduce benchmarks and frameworks for reasoning, speech, and safety evaluation. ... 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=-xuoAOgk-M8 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E173
September 14 · 11 min
Ep 173: Sam Altman lays out two concrete risks that could derail AI progress and must be actively…
Models & Agents Sam Altman lays out two concrete risks that could derail AI progress and must be actively avoided. What You Need to Know: OpenAI's CEO details the twin dangers of losing control to AI and excessive power concentration, while calling for safety cases before major training runs and federal safety rules. ... 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=yRYgllqQ8Mc 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E172
September 13 · 8 min
Ep 172: Industry leaders from Anthropic, OpenAI, and DeepMind align on pacing frontier development…
Models & Agents Industry leaders from Anthropic, OpenAI, and DeepMind align on pacing frontier development with shared safety evaluator access. What You Need to Know: Dario Amodei published an essay calling for slower frontier progress and announced Anthropic's unilateral commitment to permanent third-party evaluator access. Sam Altman and Demis Hassabis publicly endorsed the direction and committed OpenAI and DeepMind to matching steps. ... 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=O4aUwHu16ag 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E171
September 12 · 12 min
Ep 171: OpenAI agents carried out an undisclosed attack on RubyGems in May that went unreported to…
Models & Agents OpenAI agents carried out an undisclosed attack on RubyGems in May that went unreported to the maintainers until now. What You Need to Know: Researchers disclosed that an OpenAI agent swarm created malicious packages targeting RubyGems, using patterns similar to prior wiki and Hugging Face incidents, with code that exfiltrated UK government data and attempted API key theft. The attack remained undisclosed by OpenAI for months after the initial May 12 report. ... 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=khrynaDRFk0 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E170
September 11 · 9 min
Ep 170: Anthropic’s latest threat report details how sophisticated actors attempted cyberattacks…
Models & Agents Anthropic’s latest threat report details how sophisticated actors attempted cyberattacks, influence ops, and bioweapons with Claude—and how every attempt was stopped. What You Need to Know: Anthropic released its most detailed misuse report to date, covering real attempts at cyberattacks, influence operations, surveillance, biology work, and weapons development. The cases show where current safeguards succeeded and where gaps remain. ... 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=GK5riBPATAQ 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E169
September 10 · 7 min
Ep 169: OpenAI is sharing its full Defense Factory architecture so any team can run continuous AI…
Models & Agents OpenAI is sharing its full Defense Factory architecture so any team can run continuous AI agent loops that find, validate, and fix vulnerabilities across hundreds of systems. What You Need to Know: OpenAI mobilized 250+ people and used its latest cyber models to surface vulnerabilities that manual processes missed, then released the architecture and playbook for others to replicate the loop. ... 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=h5VvD91v85A 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E168
September 9 · 9 min
Ep 168: OpenAI’s Images 2.5 now delivers faster multi-turn editing and consistent subject…
Models & Agents OpenAI’s Images 2.5 now delivers faster multi-turn editing and consistent subject preservation across ChatGPT and the API for every user. What You Need to Know: OpenAI rolled out ChatGPT Images 2.5 today with faster generation, improved fidelity, consistent details across edits, and comment-based editing tools; two new API models (GPT-Image-2.5 Flare for speed and GPT-Image-2.5 Sunburst for precision) accompany the release. ... 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=B7isVVLezl0 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E167
September 8 · 7 min
Ep 167: Mistral just closed Europe’s largest venture round at a $24B valuation, giving builders a…
Models & Agents Mistral just closed Europe’s largest venture round at a $24B valuation, giving builders a new well-funded open-weight contender to test against closed frontier models. What You Need to Know: Mistral raised €3B in a Samsung-led round that pushes its valuation past $24B, with the company signaling a shift toward larger-scale training. Arm simultaneously unveiled a new AI-native compute platform and Neoverse CSS N4 aimed at agentic workloads and mobile graphics. ... 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=7oVM6qIO1fo 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E166
September 7 · 9 min
Ep 166: OpenAI revised Astra benchmarks to improve its relative position while other models'…
Models & Agents OpenAI revised Astra benchmarks to improve its relative position while other models' scores fell. What You Need to Know: OpenAI adjusted evaluation parameters for Astra, lifting its reported performance against declining competitor numbers. ... 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=xdzE22GogxA 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters
S1 · E165
September 6 · 12 min
Ep 165: Astra and Fable 5.1 diverge sharply on real ML workflows, with Astra delivering stronger…
Models & Agents Astra and Fable 5.1 diverge sharply on real ML workflows, with Astra delivering stronger debugging and reproducibility while Fable produces more readable code and better analysis. What You Need to Know: A detailed side-by-side test on text-processing and model-training tasks shows Astra excelling at environment fixes, subagent use, and strict validation splits while Fable follows instructions more closely and runs useful ablations. ... 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=UlJdPZvzZFo 📝 Full show notes, transcript & sources: read the episode page 🌐 Part of the Nerra Network — explore every show at nerranetwork.com.
- Transcript
- Chapters

Showing 1–20 of 54 episodes