Sebastian Raschka
Everything, newest first — across every channel. Their profile → · Subscribe
Hiding
17 October
10 October
9 October
6 October
4 October
3 October
2 October
29 September
27 September
26 September
23 September
22 September
20 September
19 September
15 September
14 September
13 September
10 September
9 September
From one piece GPT-6 Astra, Looped Transformers, and Hidden Reasoning 3 beliefs, in the piece's order there
-
Their words
So, in short, we can say that using looped transformers can improve model quality at a fixed compute budget if the model is large enough.
-
korrents.com
Computer-use capability will be a major focus of LLM and coding-agent development in the coming months and years.Their words
I expect the upcoming months (or years) also to be an era of computer use refinement on both the LLM and the agent harness layer.
-
Their words
GPT-6 Astra (and likely any LLM in the foreseeable future) is still a reasoning model.
8 September
6 September
2 September
1 September
30 August
29 August
26 August
22 August
-
korrents.com
Coding something from scratch is one of the best ways to truly understand how it works.Their words
So this one will not be coding-related, but coding from scratch is actually a very, very useful technique because it really helps you understand how something is implemented.
How Claude Watermarks AI-Generated Textmagazine.sebastianraschka.com
15 August
From one piece Building an AI Text Detector From Scratch 2 beliefs, in the piece's order there
-
Their words
the score is the classifier's estimated probability for the AI-generated class based on its training distribution. However, we shouldn't interpreted it as a general probability that the text was written by AI.
Building an AI Text Detector From Scratchmagazine.sebastianraschka.com
-
korrents.com
AI text detectors and the LLMs they try to catch are locked in a permanent cat-and-mouse arms race.Their words
AI checkers are essentially a cat-and-mouse game. AI checkers may learn to detect a certain pattern that is indicative of AI-generated content. Then, the next LLM may incidentally or deliberately not exhibit that pattern and avoid detection. The AI checker then has to be updated to detect said LLM, and so forth.
Building an AI Text Detector From Scratchmagazine.sebastianraschka.com
12 August
11 August
7 August
28 July
26 July
25 July
18 July
From one piece Controlling Reasoning Effort in LLMs 3 beliefs, in the piece's order there
-
korrents.com
Pushing an LLM's reasoning budget higher eventually yields diminishing returns and becomes uneconomical.Their words
This saturation can be seen more clearly for the GPT 5.6 Sol model, which also shows that increasing reasoning budgets can become uneconomical at some point.
Controlling Reasoning Effort in LLMsmagazine.sebastianraschka.com
-
Their words
a smaller model at a higher reasoning effort can sometimes reach a similar score as a larger model at a lower reasoning effort
Controlling Reasoning Effort in LLMsmagazine.sebastianraschka.com
-
Their words
These <think> and </think> tags are cosmetic with respect to reasoning ability. They do not make the model reason, and they are not required to achieve good reasoning performance.
Controlling Reasoning Effort in LLMsmagazine.sebastianraschka.com
27 June
From one piece Using Local Coding Agents 3 beliefs, in the piece's order there
-
Their words
untrusted repos should be treated as hostile by default because they can steer the agent toward reading files, running commands, or sending data through approved tools.
-
korrents.com
When an AI lab builds both a model and a coding harness, the model is optimized for that lab's own harness first.Their words
if an LLM developer also develops a coding harness, it is somewhat safe to assume that their model is optimized for their own harness first
-
korrents.com
Proprietary AI services will become more restrictive over time, making open-weight alternatives valuable as a backup.Their words
proprietary services may become more restrictive over time, and it's maybe a good idea to be comfortable with open-weight alternatives as a backup.
-
Usesrcmnd.app
CodexTheir words
I have to admit that I still primarily alternate between Codex and Claude Code as my daily drivers, for now
-
Usesrcmnd.app
Claude CodeTheir words
I have to admit that I still primarily alternate between Codex and Claude Code as my daily drivers, for now
-
Likedrcmnd.app
OllamaTheir words
I am not affiliated with any of the tools mentioned in this article, but one nice thing about Ollama is that they also optionally support open-weight models hosted in the cloud
+ 4 more
-
Usesrcmnd.app
Qwen3.6Their words
Here, I am primarily going to use Qwen3.6 with the Qwen-Coder coding client.
-
Usesrcmnd.app
Qwen-CodeTheir words
Here, I am primarily going to use Qwen3.6 with the Qwen-Coder coding client.
-
Recommendsrcmnd.app
North Mini CodeTheir words
Alternatively, if you don't want to use Qwen3.6, Cohere's North Mini Code is probably the most interesting, capable alternative at this size class right now.
-
Recommendsrcmnd.app
MLXTheir words
I highly recommend using *-mlx versions of models working on Macs (if available).
26 June
15 June
6 June
-
Their words
In 2026, long-context efficiency is king as more and more LLMs get plugged into agent harnesses (OpenClaw etc.), which requires working with longer and longer contexts.
16 May
From one piece Recent Developments in LLM Architectures: KV Sharing, mHC, and Compressed Attention 2 beliefs, in the piece's order there
-
korrents.com
Larger language models already have enough capacity that extra per-layer embedding tricks add little benefit to them.Their words
However, larger models already have sufficient capacity where these extra embeddings may not help that much.
-
Their words
As reasoning models and agent workflows keep more tokens around (for longer), KV-cache size, memory traffic, and attention cost quickly become the main constraints, and LLM developers are adding a growing number of architecture tricks to reduce those costs.
18 April
10 April
7 April
4 April
22 March
25 February
24 January
30 December 2025
21 December 2025
20 December 2025
17 December 2025
3 December 2025
4 November 2025
5 October 2025
14 September 2025
6 September 2025
9 August 2025
21 January 2025
20 January 2025
13 January 2025
Nothing matches.
What is a korrent?
A korrent is a belief a person has stated in their own words: one sentence stating the claim, backed by a quote and a source, kept at korrents.com.
Under a name here, the quoted block is what they actually said. The korrent beneath it is the claim those words support, in korrents' wording — tap it to see the record, its source, and who else holds it.
Nobody here wrote their own korrents. They are compiled from public statements, and a person can change their mind, which is recorded too.
About the English under a post
Some people here publish in a language other than English. Where they do, this site shows a machine translation beneath the post, in this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the serif above is exactly what the person published, and it is what to quote them on. A translation can be wrong in ways that matter, especially about tone.
Only the post's own words are translated. A quoted post, a linked article and a belief on korrents.com are left in their original language.