TLDRocket
Sign in

Custom Kernels for All from Codex and Claude

Hugging Face

Codex and Claude generated production-ready CUDA kernels for a video diffusion pipeline and a language model using a new agent skill that packages GPU optimization knowledge. The RMSNorm kernel achieved 1.88x speedup on isolated benchmarks for LTX-Video and 1.94x for Qwen3-8B, translating to 6% and measurable end-to-end improvements when combined with torch.compile. Users can now install the skill, prompt an agent to build optimized kernels, benchmark them against PyTorch baselines, and publish pre-compiled binaries to the HuggingFace Kernel Hub for one-line loading by others.

Read more related to this at Hugging Face
Codex Claude Qwen3-8B HuggingFace LTX-Video

Related stories

ByteDance Seed and Tsinghua AIR Introduces CUDA Agent: A Large-Scale Agentic RL System for CUDA Kernel Generation MarkTechPost · 3 weeks ago · 45 Microsoft’s SkillOpt Shows Optimized Agent Skill Artifacts Transfer Across Model Scales and Between Codex and Claude Code Harnesses MarkTechPost · 1 month ago · 34
How NVIDIA engineers and researchers build with Codex OpenAI · 4 months ago · 26
AI Agents AI Code Generation

The daily briefing

Every AI story that matters, in your inbox by 8am.

TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.