← Latest products

Warp

Run DeepSeek v4.1 Flash with 5 GB of RAM at 3.77 tok/s. Run the full 2.78-trillion-parameter Kimi K3 model, DeepSeek V4.1 Flash or GLM-5.3-Flash beyond available RAM by streaming activated weights directly from NVMe. A dependency-free, embeddable C inference engine.

Open source

2,529 stars · GitHubexternal site

Visit github.comexternal site

Reactions (2)

Similar products

by description similarity

  1. I shrank DeepSeek V4 Flash to 57GB and it wrote a compiler on my Mac

    huggingface.co

    Text Generation model. Download, explore files and run it on Hugging Face.

    Hacker News21 pts

  2. RunAnywhere (YC W26)

    github.com

    Faster AI Inference on Apple Silicon

    Hacker News240 pts

  3. sllm

    sllm.cloud

    Split a GPU node with other developers, unlimited tokens

    Hacker News188 pts

  4. Cactus Needle 3

    cactuscompute.com

    8-29MB automation models can match DeepSeek V4 Flash

    Hacker News537 pts

  5. A new engine to run Kimi K3 on a laptop

    news.ycombinator.com

    Kimi K3 has 2.78 trillion parameters and ships as 1.42 TB of weights.

    Hacker News7 pts