fal-ai / diffusion-speedrunLinks
Focused on fast experimentation and simplicity
☆74Updated 6 months ago
Alternatives and similar repositories for diffusion-speedrun
Users that are interested in diffusion-speedrun are comparing it to the libraries listed below
Sorting:
- ☆47Updated 4 months ago
- ☆23Updated last year
- supporting pytorch FSDP for optimizers☆82Updated 6 months ago
- ☆24Updated last month
- WIP☆93Updated 10 months ago
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆127Updated last year
- ☆27Updated last year
- Tiny re-implementation of MDM in style of LLaDA and nano-gpt speedrun☆52Updated 3 months ago
- ☆24Updated last year
- Synthetic Alphabet Dataset☆19Updated 2 months ago
- research impl of Native Sparse Attention (2502.11089)☆54Updated 4 months ago
- ☆30Updated 8 months ago
- ☆78Updated 11 months ago
- realtime latent world model inference demo☆46Updated 7 months ago
- ☆20Updated 8 months ago
- My take on Flow Matching☆63Updated 5 months ago
- ☆33Updated 5 months ago
- σ-GPT: A New Approach to Autoregressive Models☆65Updated 10 months ago
- ☆33Updated 7 months ago
- Simple implementation of muP, based on Spectral Condition for Feature Learning. The implementation is SGD only, dont use it for Adam☆80Updated 10 months ago
- Train VAE like a boss☆281Updated 8 months ago
- ☆28Updated 6 months ago
- Just another reasonably minimal repo for class-conditional training of pixel-space diffusion transformers.☆106Updated 3 weeks ago
- Implementation of the proposed Spline-Based Transformer from Disney Research☆97Updated 7 months ago
- Collection of autoregressive model implementation☆85Updated 2 months ago
- Lightweight package that tracks and summarizes code changes using LLMs (Large Language Models)☆34Updated 3 months ago
- ☆34Updated 9 months ago
- Implementation of Gradient Agreement Filtering, from Chaubard et al. of Stanford, but for single machine microbatches, in Pytorch☆25Updated 5 months ago
- Official implementation of the paper: "ZClip: Adaptive Spike Mitigation for LLM Pre-Training".☆127Updated 2 weeks ago
- an open source reproduction of NVIDIA's nGPT (Normalized Transformer with Representation Learning on the Hypersphere)☆101Updated 3 months ago