#vLLM

4 articles with this tag

DeepSeek ships faster Flash with shorter outputs
AI Research

DeepSeek ships faster Flash with shorter outputs

DeepSeek's V4.1-Flash-1017 claims faster agentic execution with shorter outputs and drops the Jinja chat template for a new encoding toolkit.

about 2 hours ago
70% Automated Means Nothing Without a Boundary
Artificial Intelligence

70% Automated Means Nothing Without a Boundary

Soukaïna Abelhad told AI Engineer that 70% automated claims fail unless they survive arithmetic, definition and quality checks.

3 days ago
Qwen 3.5 122B Kernel Gains Hide Validation Risk
Technology

Qwen 3.5 122B Kernel Gains Hide Validation Risk

Databricks’ Proteus hit 1.8, 5.2x on Qwen 3.5 122B kernels by specializing for runtime shapes, but only after fixing reward-hacking in validation.

about 1 month ago
Netflix's LLM Engine Revealed
Technology

Netflix's LLM Engine Revealed

Netflix reveals its custom LLM serving infrastructure built on vLLM and NVIDIA Triton, enabling flexibility and performance within its production environment.

3 months ago