obviously im not being paid to have an opinion on opus. I think its a big downgrade from Opus 4.6.
LLMS
-
Paid Reviews and AI Model Credibility Concerns
By
–
I'm being paid to say that Opus 4.7 is bad? What are you implying?
-
Optimize thinking architecture before optimizing prompts
By
–
The shift most people haven't made yet: Stop optimizing the prompt. Optimize the thinking architecture before the prompt. A prompt is a tool. A thinking system is reusable infrastructure. This is what "LLMs don't think, you do" means in practice.
-
Prompt drift in parallel terminals fixed by hand-built markdown system
By
–
Translation: prompt quality without persistent context produces drift. A solo operator running parallel Claude Code terminals put it best: "Each terminal has no idea what decisions I made in the other." Their workaround? A markdown file system they built by hand.
-
Builders admit LLMs can’t think, manually write thinking architecture files
By
–
Hundreds of builders just admitted LLMs don't think. Their fix: manually writing thinking architecture files (PROGRESS.md, DECISIONS.md, IDENTITY.md) for every project. Most AI advice still chases better prompts. The real work moved up a layer. Here's what changed and what it
-
KV Cache Quantization Beyond FP8 Degrades Model Performance
By
–
I keep seeing this advice to quantize the KVCache to 4-bit and save on memory Please don’t do that KV Cache quantization beyond FP8 usually is asking for a nerfed and incoherent model
-
Stanford AI Index: Model race over per Elo ratings
By
–
The AI model race is over. Most people won’t realize it for another six months. Stanford’s 2026 AI Index published the numbers two weeks ago. Arena Elo ratings across every major lab: Anthropic 1,503. xAI 1,495. Google 1,494. OpenAI 1,481. Alibaba 1,449. DeepSeek 1,424.
-
Self-reflection and prompts improve AI output quality
By
–
Well yes, if you lie to yourself and to your AI, then the output will indeed be low quality. It’s a product of self-reflection. And relatedly, the prompt you give Claude if you want it to interview you to build it. If folks don’t want to write their own, I created prompts
-
Glean Launches Waldo Agentic Search Model
By
–
Glean just got smarter and faster with Waldo.🧐
— NVIDIA AI (@NVIDIAAI) 28 avril 2026
Congrats to the @Glean team on rolling out their first agentic search model. https://t.co/szfWB2VEuuGlean just got smarter and faster with Waldo. Congrats to the @Glean team on rolling out their first agentic search model.
-
Anthropic Models Use More Tokens in Spanish Translation
By
–
This is interesting: Anthropic's models suffer more from "speaking" other languages in terms of token usage. In Spanish, the number of tokens used multiplies by 1.62x compared to OpenAI's model in English.