This week in AI: a 2.8T open-weight model topped a frontier code benchmark, Grok landed inside Excel, and Anthropic quietly closed the ZDR loophole on Fable 5. Building right now? The ground shifts weekly — worth an hour to recheck your stack.
#AI#IndieHackers#TechNews
AI pricing war: Grok 4.5 at $2/$6 per million tokens vs Claude Opus 4.7 at $5/$25 — builders are posting ROI comparisons showing it holds up on coding work. Model choice is now a cost decision, not just a capability one.
#Grok#AI#BuildInPublic
https://t.co/az68A3DKP6
Claude Fable 5 requires 30-day data retention and isn't available under Zero Data Retention — even orgs with an existing ZDR deal don't get it for Fable 5. Handling client data? Check this first.
#AI#DataPrivacy#Compliance
https://t.co/OQYqEn9S78
Grok now lives inside Excel. Ask questions in plain English, get real formulas written into the formula bar, run "what if" scenarios — no exports, no copy-pasting into a chat window.
#Grok#Excel#AITools
https://t.co/G43e5OzQ9r
Moonshot AI released Kimi K3 — 2.8T parameters, the largest open-weight model ever. It topped the Frontend Code Arena benchmark, beating Claude Fable 5, GPT-5.6 Sol and GLM-5.2. Fully open-source by end of July.
#AI#OpenSource#IndieHackers
https://t.co/oNJQpnZRNz