Understanding the Memory Breakthrough in Opus 4.5
In Machines we Trust
In Machines we Trust
4.3 • 6 Ratings
🗓️ 26 November 2025
⏱️ 8 minutes
🧾️ Download transcript
Summary
Opus 4.5 features a memory redesign built for sustained reasoning. Anthropic highlights huge gains in multi-turn understanding and recall consistency. We dive into what this means for AI reliability.
- Get the top 40+ AI Models for $20 at AI Box: https://aibox.ai
- AI Chat YouTube Channel: https://www.youtube.com/@JaedenSchafer
- Join my AI Hustle Community: https://www.skool.com/aihustle
See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.
Transcript
Click on a timestamp to play from that location
| 0:00.0 | The weight is over. Dive into Audubles' most anticipated collection, the best of 2025, featuring top audiobooks, podcasts, and originals across all genres. Our editors have carefully curated this year's must-listens, from brilliant hidden gems to the busiest new releases. |
| 0:23.7 | Every title in this collection has earned its spot. |
| 0:28.3 | This is your go-to for the absolute best in 2025 audio entertainment. |
| 0:31.6 | Whether you love thrillers, romance, or non-fiction, |
| 0:33.6 | your next favorite listen awaits. |
| 0:38.8 | Discover why there's more to imagine when you listen at audible.com slash best of the year. |
| 0:45.9 | Anthropic has just announced Opus 4.5, which is the latest version of its flagship model. |
| 0:55.8 | It's the last of Anthropics 4.5 series model that's going to be released, and it's following the launch of Sonnet 4.5, which happened back in September. |
| 1:01.5 | And we had Haiku 4.4.5 in October. So this is kind of the big model everyone has been waiting for out of Anthropic. I think as to be expected, the new version of Opus has, you know, |
| 1:07.6 | basically state-of-the-art performance on a bunch of different benchmarks, including the SWE benchmark for software engineers. This is in, you know, basically state-of-the-art performance on a bunch of different benchmarks, |
| 1:11.4 | including the SWE benchmark for software engineers. |
| 1:14.3 | This is incredibly important, also terminal bench. |
| 1:17.3 | These are very important, as Claude CodCode has become the favorite AI tool among most developers, |
| 1:22.2 | although it's starting to get some competition from GROC and from Gemini and Open AI. |
| 1:26.5 | They all kind of battle in there, but it seems |
| 1:28.3 | to be holding the number one spot. The tool use, which is the TAU2 benchmark, the MCP Atlas, |
| 1:36.9 | and general problem solving on ARC AGI 2 and GPQA Diamond. So it's doing really good on a lot of the benchmarks. |
| 1:45.7 | Today on the podcast, I want to talk about some of the changes that we can expect to see because of this. |
| 1:51.8 | But before we get into that, I wanted to mention if you want to test all of the AI models I talk about on the show, I'd love for you to try AIbox. |
| 1:58.0 | It's my own startup where you get access to over 40 different |
| 2:01.7 | models. You can compare them side by side, everything from Anthropic to OpenAI to Google Gemini |
| 2:06.9 | to GROC and a ton of different image models and for audio you have 11 labs. So it's all in there |
... |
Please login to see the full transcript.
Disclaimer: The podcast and artwork embedded on this page are from In Machines we Trust, and are the property of its owner and not affiliated with or endorsed by Tapesearch.
Generated transcripts are the property of In Machines we Trust and are distributed freely under the Fair Use doctrine. Transcripts generated by Tapesearch are not guaranteed to be accurate.
Copyright © Tapesearch 2026.