Brieftide
Topic hub
Model Serving Systems
Covers algorithms and systems for routing, scheduling, and optimization of inference workloads and multi-agent request flows for large models.
71 briefs
Model Serving Systems · Page 3
- AgentsView: Set a custom model price for Claude Fable 5The Brieftide
- Anthropic and Stanford: why larger LLMs learn rare skillsThe Brieftide
- Thousand Token Wood: five labs' small-model finance dramaThe Brieftide
- Mellum2 by JetBrains: 12B MoE model with 2.5B active paramsThe Brieftide
- MagenticLite, MagenticBrain & Fara1.5: Microsoft Research releaseThe Brieftide
- OlmoEarth v1.1 release: Up to 3× cheaper satellite AIThe Brieftide
- GridSFM release: Microsoft's model solves AC‑OPF in millisecondsThe Brieftide
- AWS building blocks for foundation model training and inferenceThe Brieftide
- GRASP: Gradient Planning for World Models at Long HorizonsThe Brieftide
- MIT pruning method trims models during training, reduces computeThe Brieftide
- Algorithms: Dimitris Bertsimas Killian Lecture on life andThe Brieftide
- Hugging Face and NVIDIA: build domain embeddings in a dayThe Brieftide
- Holotron-12B release: 12B agent for high-throughput computeThe Brieftide
- Async RL libraries: 16 open-source projects from Hugging FaceThe Brieftide
- Modular Diffusers: Hugging Face releases composable v1The Brieftide
- Mixture of Experts (MoE) in Transformers: Hugging Face guideThe Brieftide
