H100
NVIDIA's high-performance GPU, used as a benchmark to discuss the 30x efficiency improvement of the upcoming b100s for inference.
Common Themes
Videos Mentioning H100
The 10,000x Yolo Researcher Metagame — with Yi Tay of Reka
Latent Space
The H100 GPU series experienced major delays and supply chain issues, which significantly impacted Reka AI's initial training runs due to unreliable hardware.
DeepSeek V3, SGLang, and the state of Open Model Inference in 2025 (Quantization, MoEs, Pricing)
Latent Space
A GPU that is insufficient for serving the DeepSeek V3 model due to memory constraints (requiring 640 GB for weights alone, plus KV cache).
E167: Google's Woke AI disaster, Nvidia smashes earnings (again), Groq's LPU breakthrough & more
All-In Podcast
Nvidia's GPU model, mentioned as part of the hardware being used to augment data centers.
![2024 in Post-Transformer Architectures: State Space Models, RWKV [Latent Space LIVE! @ NeurIPS 2024]](https://i.ytimg.com/vi/LPe6iC73lrc/maxresdefault.jpg)
2024 in Post-Transformer Architectures: State Space Models, RWKV [Latent Space LIVE! @ NeurIPS 2024]
Latent Space
A high-performance GPU mentioned as an example where traditional RNNs cannot achieve high utilization due to their sequential nature.
llm.c's Origin and the Future of LLM Compilers - Andrej Karpathy at CUDA MODE
Latent Space
Nvidia's high-performance GPU, capable of training large AI models like GPT-2 efficiently.
Open Source AI is AI we can Trust — with Soumith Chintala of Meta AI
Latent Space
NVIDIA's high-performance GPU, used by Meta to measure its aggregate computing capacity.
Building an open AI company - with Ce and Vipul of Together AI
Latent Space
A high-end NVIDIA GPU mentioned in the context of inference performance and comparisons with other systems.
Beating GPT-4 with Open Source Models - with Michael Royzen of Phind
Latent Space
NVIDIA's high-performance GPU, mentioned in the context of their FP8 implementation for deep learning inference.
Ep 18: Petaflops to the People — with George Hotz of tinycorp
Latent Space
The cost of an H100 box is mentioned as a benchmark for comparison with Tiny Corp's offerings.
SF Compute: Commoditizing Compute
Latent Space
A high-end NVIDIA GPU model frequently discussed in the context of supply, demand, and pricing in the GPU cloud market.
The Shape of Compute (Chris Lattner of Modular)
Latent Space
A newer NVIDIA GPU architecture for which Modular added support, improving performance and features.
Aravind Srinivas: Perplexity CEO on Future of AI, Search & the Internet | Lex Fridman Podcast #434
Lex Fridman
NVIDIA's high-performance GPU, used as a benchmark to discuss the 30x efficiency improvement of the upcoming b100s for inference.