ThunderAgent: 2x Faster Agentic Inference for Synthetic Data Generation
togethercompute · x · 2026-07-30
Together AI introduced ThunderAgent, a high-throughput system designed for large-scale agentic inference. By introducing a novel program abstraction for scheduling agentic LLM requests, it effectively eliminates KV cache thrashing.
Key Performance Results:
- Achieves >2× single-node throughput with roughly 10× lower P50 latency at high concurrency.
- Delivers a 2.4× speedup on an 8-node cluster, demonstrating near-linear throughput scaling from 16 to 64 GPUs.
Engineering Integration:
- Acts as a drop-in solution requiring only a single programid field, plugging into existing engine configs like KV offloading and speculative decoding.
- Format-agnostic by design, supports OpenAI chat completions today, and has already been adopted by SkyRL and NVIDIA Dynamo.
The research was accepted as a Spotlight paper at ICML 2026.
Related event: Together AI Unveils ThunderAgent, Doubling Agent Inference Throughput(7 posts)→
More from Infra
- NVIDIA launches RTX Spark for local AI on Windows PCs — ostrisai · 2026-10-16
- SkyRL ships Day-0 RL support for Beam with +62% LoRA decode throughput in vLLM — vllm_project · 2026-10-16
- DFlash2 delivers up to 3.03x decode-only speedup for Beam across five benchmarks — vllm_project · 2026-10-16
- vLLM's Beam serving kernels cut GB200 forward time from 8.15 to 6.47 ms at TP8 batch 128 — vllm_project · 2026-10-16
- Reflection's 501B MoE Beam hits 463 tok/s on 8x GB200 with NVFP4 at day 0 — zhyncs42 · 2026-10-16
- Gemma 4 31B QAT on one 24GB card: 46 tok/s plain, ~80 with draft model on RTX 4090 — Subject_Poet_3861 · 2026-10-16