Pinned
- Why isn’t anyone talking about deepseek’s size? The new Flash is only ~300B parameters, yet it’s performing on par with models several times larger. Deepseek is doing some insane work on efficiency Can’t wait for V4 Pro.
- I don’t get it, is it just t3 code/opencode but worse?We just launched Xirp, a vendor-neutral agentic development environment. One place to manage agent sessions across @ClaudeDevs, @GeminiApp CLI, and @OpenAI Codex. 1,300+ @Spotify engineers already use it. Now it's available for you to try. Learn more at xirp.spotify.com.
- I don't really understand the point of this. I can already prompt codex (or any other agent) to do something, and when i come back, its done too. Is there a point in using this over any other ai agent?
