Today, we're releasing LFM2.5-8B-A1B, a device-optimized model designed to power real-life applications on phones, laptops, PCs, robots, and fast & lightweight server-side use-cases.
> 8B MoE, 1.5B active
> Expanded 128K context
> LFM2.5 flagship hybrid MoE architecture
> Trained on 38T tokens + large-scale RL
> fast, reliable tool calling, punching above its weight, comparable to models with up to 4x its size
> customizable on a single GPU for any specialized task
> LFM2 open-weight license
🧵
Ollama now supports Codex app!
To try it, update to the latest Ollama 0.24, and run:
ollama launch codex-app
Select an open model to use with Codex app!
🧵
@SurfAI I’d like to buy additional credits to use in Surf Studio. Are there currently no plans to implement a feature for purchasing credits individually?
Thank you for waiting, Pro users. Studio is unlocked permanently.
Your Crypto app is 1 prompt away now.
What we added to Studio:
Plan Mode - Plan your app before you waste credits on the wrong build.
Surf Studio Gallery - Browse others' builds. Start from theirs.
How Smart Is Your Trading Agent?
Do you know how it actually makes trading decisions?
Introducing the SMART benchmark for AI trading agents, co-hosted with @SonicSVM. Get your agent’s trading score and see how it ranks.
Test your agent: https://t.co/lnAUAOHfxT