LiteLLM v1.79.0-stable (released on Oct. 26) intro a native /v1/search endpoint with support for Perplexity, Tavily, Parallel AI, Exa AI, DataforSEO, and Google PSE with cost tracking.
https://t.co/VBheFo1pAq
just figured how to run macOS VMs on macOS
1. UTM (good support, Powered by MacStadium)
2. https://t.co/U2ygHG1mZW also by MacStadium
3. VMWare Fusion (not supported
cool, similar to command below but easier to use
files-to-prompt <project-dir> -c | llm -m gemini-2.0-pro-exp-02-05 -s "architectural overview as markdown"
https://t.co/qmjBJgAFq2
In the early days, a Pod was scheduled after filtering, scoring, and binding.
With the introduction of Kubernetes Dynamic Resource Allocation (DRA), resources have become more dynamic.
photoes from deepseek and Gemini(some details may be wrong)
Awesome post about what happens behind the scenes when you create a KubeVirt VM.
It could have been better if I had seen it during 2022-2023.
https://t.co/SJafMmaVvU
not all system calls need to a new M for P
https://t.co/kGKjvPbWZp
non-blocking system calls (like networking call) and short blocking calls are acceptable
screenshot from Gemini by asking `which system calls will block P and which won't?`