HN Explorer

OsamaJaber

↗ HN profile [ 63.4 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. Free Cached Input (runinfra.ai)
    2 points 0
  2. Ornith 1.5 122B at 302tps (runinfra.ai)
    2 points 0
  3. The fastest DeepSeek v4.1 flash 1017 – Day 0 (runinfra.ai)
    3 points 0
  4. Is this the best inference for agents? (runinfra.ai)
    2 points 1
  5. DeepSeek v4.1 Flash at 378 tok/s 99.7% Cache hit rate (runinfra.ai)
    1 points 0
  6. GLM 5.3 flash on AMD GPUs 670 tok/s (twitter.com)
    2 points 1
  7. Agent loops stop re-reading your codebase on every turn (runinfra.ai)
    1 points 0
  8. GLM 5.3 Flash faster and cheaper (runinfra.ai)
    1 points 0
  9. The fastest and cheapest GLM 5.3 Flash endpoint (runinfra.ai)
    2 points 0
  10. DeepSeek V4 Pro at 207 tok/s with the full 1M context, no quantization (runinfra.ai)
    3 points 0
  11. Fastest Inference in MENA (runinfra.ai)
    2 points 0
  12. DeepSeek V4 Flash at 278 tok/s, full precision, no quantization (runinfra.ai)
    10 points 4
  13. The fastest full-precision Nemotron 3.5 Lightning endpoint (540 tok/s) (runinfra.ai)
    4 points 0
  14. With software alone, one B200 beats the LPU and gets close to Cerebras (runinfra.ai)
    11 points 1
  15. Kimi K3 2.78T on One CPU with 8GB RAM (github.com)
    4 points 0
  16. Mixture-of-Kittens: An MoE training megakernel for NVL72 (twitter.com)
    3 points 0
  17. Lossless Inference (runinfra.ai)
    4 points 0
  18. DeepSeek V4 Flash 2.98x faster, lossless (runinfra.ai)
    1 points 1
  19. DeepSeek-V4-Flash 2.98x faster on 4x B200, lossless (twitter.com)
    3 points 0
  20. 8 points 0
  21. 5 points 0
  22. What LLM Inference Costs (twitter.com)
    2 points 0
  23. Kimi k3 run on RTX 5090 (github.com)
    4 points 0
  24. Kimi k3 now runs on one consumer GPU (twitter.com)
    8 points 0
  25. The AGI Compiler "Auto" (github.com)
    2 points 0