Models & Labs
ThunderAgent Boosts Agentic Inference Efficiency
Together AI Bloghigh confidence
Why it matters
- →ThunderAgent significantly reduces latency and increases throughput in agentic inference.
- →It addresses KV cache thrashing, a major bottleneck in high-concurrency environments.
- →The system is compatible with existing inference optimizations, facilitating easy adoption.
Together AI has unveiled ThunderAgent, a system designed to enhance agentic inference for synthetic data generation. By abstracting agent workflows as programs, ThunderAgent addresses inefficiencies like KV cache thrashing, achieving up to 2.5× higher throughput on single nodes and 2.4× speedup on 8-node clusters. This approach allows for better load balancing and cache utilization, making it a valuable tool for large-scale agentic workloads. ThunderAgent is open source and integrates seamlessly with existing inference systems.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
More in Models & Labs
Models & Labsother
llama.cpp b11193 adds Snapdragon Hexagon and ROCm 10
This release quietly expands llama.cpp's hardware support to include Qualcomm's Hexagon NPU on Linux arm64, a significant step for local inference on Snapdragon devices. It also updates CUDA builds to version 13.4 and introduces ROCm 10.0 binaries, keeping the project aligned with the latest NVIDIA and AMD driver ecosystems. KleidiAI on Apple Silicon is temporarily disabled in this build, likely due to stability checks rather than a feature rollback. For developers targeting edge AI or diverse GPU stacks, this update ensures broader compatibility without requiring custom compilation.
llama.cpp Releases
© Lev SelectorModels & Labsmodels
NVIDIA NVFP4 and SoL-Pi Token Efficiency
NVIDIA introduced the NVFP4 4-bit format and SoL-Pi technology, which uses 2x fewer tokens for improved efficiency.