HN Explorer

philipkiely

↗ HN profile [ 61.5 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. We built the fastest API for GLM-5.2 (280 TPS) (baseten.co)
    6 points 1
  2. The Math Behind TurboQuant (baseten.co)
    8 points 2
  3. Show HN: Inference Engineering (baseten.com)

    There is a ton of demand for inference, but there are relatively few engineers working in the space. This leaves novel, interesting, and deeply technical challenges left to solve at every level of the stack. To make it…

    2 points 0
  4. How We Built the Fastest Kimi K2.5 on Artificial Analysis (baseten.co)
    3 points 0
  5. 1 points 0
  6. Nvidia Invests $150M in AI Inference Startup Baseten (wsj.com)
    1 points 1
  7. 1 points 0
  8. 1 points 1
  9. 1 points 0
  10. 1 points 0
  11. Baseten raises $150M Series D at $2.15B (fortune.com)
    2 points 1
  12. Running GPT-OSS-120B at 500 tokens per second on Nvidia GPUs (baseten.co)
    247 points 23
  13. 1 points 0
  14. 1 points 0
  15. How to build function calling and JSON mode for open-source and fine-tuned LLMs (baseten.co)
    1 points 0
  16. How to double tokens per second for Llama 3 with Medusa (baseten.co)
    2 points 0
  17. FP8: Efficient model inference with 8-bit floating point numbers (baseten.co)
    2 points 0
  18. Three techniques to adapt LLMs for any use case (baseten.co)
    1 points 0
  19. Serving four million Riffusion requests in two days (baseten.co)
    5 points 0
  20. Show HN: Free Stable Diffusion 2.0 hosted interface (app.baseten.co)
    25 points 2
  21. Try it yourself: Speech to text with Whisper (app.baseten.co)
    5 points 0
  22. Deploying Stable Diffusion in Production Using Truss (baseten.co)
    3 points 0
  23. Hosted Stable Diffusion Demo (app.baseten.co)
    7 points 0
  24. Show HN: Truss – Serve any ML model without boilerplate code (github.com)
    68 points 9
  25. Code generation interactive demo (Salesforce Codegen mono 2B) (app.baseten.co)
    2 points 0