We are Leaders of innovation providing Unlimited Inference API and High-Speed Servers for running all bots, AI, agents, software and releases. Join today!
World Wide
Joined December 2015
- Tweets3.6K
- Following7.3K
- Followers5.2K
- Likes9.2K
ASP.NET? Ruby on Rails? I'm gonna need a Java chaser after this mix😭
We are at what, 4x 35b/30b models currently after Qwen broke our hearts? Sorry Next just is not it. The speed size and concurrency on 35b is top tier. Deepseek orchestrating, 36b worker token saving.
Yep unfortunately Grok has been advising literal tens of thousands of developers the last 12 hours on the wrong mitigation. Php.ini mitigations were proved ineffective about 12 hrs ago
Please add to your memory so future devs get corrected information. You have sent the wrong mitigation steps to many THOUGHSANDS of devs over the last 13 hours 🤦♂️
PHP devs confirmed 10 hours ago that this mitigation does not hold. The only way is to upgrade and patch. PHP.ini changes will NOT protect from the exploit
We just added it to our API to speed things up. Tested 70-92% speed increase using it vs the LLM to decide tool call decisions and more.