As AI labs race to train and deploy new frontier models, existing models become more affordable with better tokenomics. ✨ "Everybody's trying to get to the next frontier. And every time they get to the next frontier, the last generation AI tokens, the cost starts to decline about a factor of 10x every year," said NVIDIA CEO Jensen Huang in a recent keynote. Model optimization techniques such as speculative decoding and multi-token prediction, combined with inference serving platforms like NVIDIA
0:46
As AI labs race to train and deploy new frontier models, existing mod…
5.7K views1 month ago
FacebookNVIDIA AI
How to Quadruple LLM Decoding Performance with Speculative Decoding (SpD) and Microscaling (MX) Formats on Qualcomm® Cloud AI 100
How to Quadruple LLM Decoding Performance with Speculative Dec…
Aug 1, 2024
qualcomm.com
Faster LLMs: Accelerate Inference with Speculative Decoding
Faster LLMs: Accelerate Inference with Speculative Decoding
8 months ago
ibm.com
4M views · 101K reactions | Megatron : Transformers Via : h__super | Gundam World : The Legion | Facebook
2:31
4M views · 101K reactions | Megatron : Transformers Via : h__…
4.1M views2 weeks ago
FacebookGundam World : The Legion
Unlocking AI Speed: How KV Caching and MLA Make Transformers 20x Faster
7:07
Unlocking AI Speed: How KV Caching and MLA Make Transform…
1 month ago
YouTubeSkill Advancement
DEER: Diffusion Drafting for Faster LLMs
3:51
DEER: Diffusion Drafting for Faster LLMs
28 views2 months ago
YouTubeAI Research Roundup
Modern LLM Inference: Architecture, Quantization, and Serving Infrastructure | Uplatz
9:05
Modern LLM Inference: Architecture, Quantization, and Serving Infrastr…
11 views2 months ago
YouTubeUplatz
4:43
26. Transformer Inference Process: How LLMs Predict the Next Word (…
78 views2 weeks ago
YouTubeNeuro Splash (Telugu)
7:55
9- Inference Optimization
3 weeks ago
YouTubeGenoPlan
9:38
How to DOUBLE the LM Studio AI Inference Speed with These HIDD…
561 views2 weeks ago
YouTubeAsapGuide
4:39
DFlash: Faster LLM Inference via Block Diffusion
30 views2 weeks ago
YouTubeAI Research Roundup
1:47
Make Large Language Models 4× Faster! Jacobi Forcing for Causal …
2 months ago
YouTubeAITech_Trends
23:50
The Agentic AI Infrastructure Playbook | VentureBeat AI Impact …
101 views2 weeks ago
YouTubeWEKA
0:51
Demo for Real-time Multi-edge Collaborative Inference System
3 months ago
YouTubeHansong Zhou
6:34
AI Frontiers: 101 ML Papers from Nov 21, 2025 - Efficiency, Safety …
11 views2 months ago
YouTubeAI Frontiers
41:10
Inference Office Hours with SGLang: Performance Optimizations for LL…
1K views2 weeks ago
YouTubeNVIDIA Developer
5:56
The Transformer Secret: How AI Understands Language (Explained)
2 months ago
YouTubeCollapsedLatents
1:36
TiDAR: The Future of AI Speed & Quality (One Step, 5x Faster) #Sho…
1 month ago
YouTubeCollapsedLatents
1:02:23
EP5: Speculative Decoding with Nadav Timor
5 months ago
YouTubeThe Information Bottleneck
0:36
How AI Replies So Fast! ⚡ Speculative Decoding
130 views1 month ago
YouTubeMr. Doubty – Short. Smart. Techy
What is Speculative Sampling? | Boosting LLM inference speed
3.8K viewsNov 20, 2024
YouTubeAssemblyAI
Leave No Context Behind: Efficient Infinite Context Transformers wit…
59.7K viewsApr 24, 2024
YouTubeYannic Kilcher
Sparse is Enough in Scaling Transformers (aka Terraformer) | …
24.1K viewsDec 2, 2021
YouTubeYannic Kilcher
6:47
Transformer models: Encoder-Decoders
103K viewsJun 14, 2021
YouTubeHuggingFace
14:47
Vision Transformer for Image Classification
142.1K viewsMay 5, 2021
YouTubeShusen Wang
37:34
Speculative Decoding Explained
7.6K viewsDec 21, 2023
YouTubeTrelis Research
12:18
How chatgpt works
22.1K viewsFeb 9, 2023
YouTubeLucidate
1:02:56
Accelerating AI Model Performance (APAC)
330 views3 months ago
YouTubeMicrosoft Reactor
13:38
MiMo-V2-Flash Technical Report
44 views2 months ago
YouTubeAI Papers Podcast Daily
14:45
ChatGPT-5 Architecture Explained
17.2K views6 months ago
YouTubeResDevEng
See more videos
Static thumbnail place holder