#ttft

The latest AI and developer news about #ttft, with the original source and a short summary.

vLLM Introduces Push-Based KV Transfer to Cut TTFT by Up to 3x in Disaggregated Inference

The new NIXL connector-based push method reduces Time To First Token by up to 3x and Time Per Output Token by up to 30% compared to pull-based approaches.

·2·2026.10.14 09:00

NVIDIA Dynamo Unveils Vision Encoder Disaggregated Serving Technology

NVIDIA Dynamo has unveiled a serving architecture that separates vision encoders to reduce the TTFT of multimodal LLMs by up to 93%.

·1·2026.09.11 07:00

KV crosses the DC

Kimi Linear makes cross-datacenter Prefill/Decode disaggregation practical, lowering costs.

·2026.04.19 01:00

Scaling Managed Agents: Separating the Brain from the Hands

By separating session, harness, and sandbox, Managed Agents gained better scalability and security.

·2·2026.04.09 09:00

RNGD Outpaces RTX Pro 6000 with Latest SDK

With its latest SDK, RNGD boosted the number of users per 15kW rack to up to 2x that of the RTX Pro 6000.

·1·2026.04.02 09:00