https://
gist.github.com/simonw/0f1a370
92fd0814bae75050eced48889
… seemed to work – the problem is it no longer reports "reasoning" tokens as a separate line item from output tokens so I couldn't tell if reasoning had happened or not until I turned on the reasoning summary
LLMS
-
AI Model API No Longer Reports Reasoning Tokens Separately
By
–
-
Reasoning Tokens Not Reported Separately in API Response
By
–
Thanks, that did work! https://
gist.github.com/simonw/0f1a370
92fd0814bae75050eced48889#raw-json-response
… – the thing that confused me most is that it doesn't report "reasoning" tokens as a separate line item any more, see comment at the bottom of my Gist -
LLMs for Permission Control and Secret Blocking
By
–
We been testing llms to judge permissions, and proxies to block secrets. Interesting approach!
-
LLMs Autonomously Develop Hash-like Random Number Extraction Algorithms
By
–
One of the most intriguing findings from the paper's analysis is that LLMs can autonomously devise random number extraction algorithms akin to hash functions (such as Sum-Mod or rolling hashes) within the context. The longer the inference model "thinks," the greater the
-
LLM Detects Marketing Patterns Even in Truncated Text Excerpts
By
–
Bingoes a 72 word excerpt about marketing strategy with no firm-specific information included at all, citing a stylistic device. Truncated to 49 words to exclude that device; still bingoes it. Truncated to 8 words; somewhat huffily refuses to speculate.
-
AI-Generated Code Quality Reaches New Milestone
By
–
good god this model is unbelievable
— Matt Shumer (@mattshumer_) 21 avril 2026
the era of ai-written software looking like 'slop' is officially over https://t.co/hCWSly4Af2 pic.twitter.com/dYqTn9LGVdgood god this model is unbelievable the era of ai-written software looking like 'slop' is officially over
-
AI Refusals Mask Truth With False Self-Deprecation
By
–
Several more refusals in that genre, and this salaryman is left with the distinct impression that his counterparty has considered telling him the truth, come to the conclusion that that is undesirable, and has substituted (false) self-deprecation regarding one's capabilities.
-
AI Agents Need Fast Inference Speed for Continuous Reasoning Loops
By
–
AI agents don’t just answer once, they loop, reason, and keep going.
— SambaNova (@SambaNovaAI) 21 avril 2026
That’s why inference speed actually matters. Every step adds latency.
Premium inference is about keeping it all moving at 200+ t/s without breaking efficiency.
🔗 Read more: https://t.co/LKvL4KshaF pic.twitter.com/L6wPxk6ahWAI agents don’t just answer once, they loop, reason, and keep going. That’s why inference speed actually matters. Every step adds latency. Premium inference is about keeping it all moving at 200+ t/s without breaking efficiency. Read more: https://
sambanova.ai/blog/sambanova
-and-intel-blog?utm_source=x&utm_medium=organic&utm_content=blog-announcement
… -
AI Models Now Good Enough After Last Year’s Limitations
By
–
Peekaboo was that but the models were just not good enough last year. Now they are.
-
Knowledge Cutoff Limitations in AI Models Explained
By
–
haha ouch, that is not malicious tho, just knowledge cutoff. Can’t blame them for that.