Logo
ResearchAudio
Search
Latest
Free tools
Community
Login
Subscribe free
Logo
ResearchAudio

Archive

DeepSeek released its AI engine but kept its storage

Oct 22, 2026

That storage lets it reuse data for 95 of every 100 tokens and keep its costs low

Read More

Same Weights, New Files: GLM-5 Scored 11.5 Points Higher

Oct 19, 2026

Changes with no test run decreased a code benchmark result from 73.4 to 63.2.

Read More

Codex Found the Right Number, Then Used the Wrong One

Oct 14, 2026

Memory alone did not fix this. Prompt rules that steer the search did.

Read More

GPT-5.4 Obeyed “No Reasoning” and Lost 28.5 Points

Oct 13, 2026

Thinking off still reasons in 99.69 percent of open answers. The fix is the question.

Read More

Your LLM server was built for one model. Token routing needs two.

Oct 12, 2026

A NeurIPS 2026 paper from Tsinghua's thu-nics group explains why token-level routing has been stuck in notebooks, and ships a serving system that makes it 2–64× faster than what researchers had been running.

Read More

Can We Still Watch AI Think?

Oct 6, 2026

Chain-of-thought provides insight into AI reasoning. New research shows where that window starts to break.

Read More

Sonnet 5.5 kept the price. Your bill could shrink.

Sep 29, 2026

Faster output, fewer tokens, and the API changes to check before switching.

Read More

The cheaper model can cost more

Sep 24, 2026

Opus 5.5 and GPT-6 Sol, explained through five diagrams about agents, caching, and the cost of finished work.

Read More

Grok 4.7 Beats a Grok 4.6 That Was Corrected in August

Sep 22, 2026

Three regressions shrank by revision. Opus 5 left the coding charts. Biology fell by design.

Read More

Your AI agent says it’s done. Is it?

Sep 18, 2026

Designing the system that assigns the work, checks the result, and knows when to stop.

Read More
Load more

ResearchAudio

Evidence-led AI briefings and free decision tools for engineers, researchers, and product builders.