zhisbug on HN Glance

[dead] 4 months ago

[dead] 12 months ago

live demo is here: https://fastwan.fastvideo.org/

[dead] 1 year ago

Pokémon is increasingly used to evaluate modern large language models, but current practices lack standardization, and depend heavily on game-specific harness. The Pokémon Red involves three major tasks—navigation, combat control and training a competitive Pokémon team. We find they come with limitations: navigation tasks are too hard, combat control is too simple, and Pokémon team training is too expensive. We address these issues in Lmgame Bench, a new framework offering standardized evaluations and initial results across diverse games.

Create a 5s 1080p Video in 4.5s with FastVideo on a Single GPU

Can LLMs play real-time games like supermario (other than Pokemon red)?

Sliding Tile Attention: A New Method That Speeds Up HunyuanVideo's Outputs by 3x

Fast Video Generation with Sliding Tile Attention

More Efficient Chain-of-Thought Reasoning Through Certainty Probing

AI Space Escape: Playing Games While Evaluting LLM Reasonsing

Efficient LLM Scheduling by Learning to Rank

FastVideo: a lightweight framework for accelerating large video diffusion models

MuxServe: Flexible Spatial-Temporal Multiplexing for Multiple LLM Serving

Consistency LLM: converting LLMs to parallel decoders accelerates inference 3.5x

Throughput Is Not All You Need: Maxing Goodput in LLM Serving via Disaggregation

Break the Sequential Dependency of LLM Inference Using Lookahead Decoding

Important and *MUST-KNOW* techniques for a 2023 LLM serving system

Fastchat-T5: 4x smaller but more powerful than Dolly-v2, commercial use ready

Alpa: Auto-parallelizing large model training and inference (by UC Berkeley)

Important and MUST-KNOW techniques for a 2023 LLM serving system