1. 27 Jun, 2025 2 commits
  2. 26 Jun, 2025 1 commit
  3. 24 Jun, 2025 1 commit
  4. 18 Jun, 2025 2 commits
  5. 17 Jun, 2025 2 commits
  6. 14 Jun, 2025 1 commit
  7. 13 Jun, 2025 4 commits
  8. 12 Jun, 2025 1 commit
  9. 11 Jun, 2025 1 commit
  10. 10 Jun, 2025 1 commit
  11. 08 Jun, 2025 1 commit
  12. 05 Jun, 2025 2 commits
  13. 04 Jun, 2025 4 commits
  14. 03 Jun, 2025 1 commit
  15. 02 Jun, 2025 3 commits
  16. 30 May, 2025 3 commits
  17. 29 May, 2025 2 commits
  18. 28 May, 2025 5 commits
  19. 27 May, 2025 1 commit
  20. 23 May, 2025 1 commit
  21. 22 May, 2025 1 commit
    • Graham King's avatar
      feat(dynamo-run): Allow setting KV cache block size (#1175) · 183f2b32
      Graham King authored
      Example:
      ```
      dynamo-run out=<engine> <model> --kv-cache-block-size 64
      ```
      
      In a distributed system this goes on the worker node and is propagated to ingress via the model deployment card.
      
      Previously hard coded to 16, which is now the default.
      
      - Load context_length from model. Closes #1172
      - Store context length and KV cache block size in Model Deployment Card #1170
      183f2b32