Robin L. de Bruijn

Robin1986

https://potionmemory.com

rldebruijn

AI & ML interests

Agent memory, long-context evaluation, benchmark validity, small open models as readers over retrieved memory.

Recent Activity

liked a model 11 days ago

Aleph-Alpha/Kolibri-1-BF16

liked a model 11 days ago

Aleph-Alpha/Kolibri-1

new activity 11 days ago

Aleph-Alpha/Kolibri-1:[Evaluation Report] Kolibri-1 FP8 on one RTX PRO 6000 (96 GB): serving numbers, and BEAM 100K from its own 262K window vs a 7K-token memory page

View all activity

Organizations

None yet

liked 2 models 11 days ago

Aleph-Alpha/Kolibri-1-BF16

Text Generation • 78B • Updated 3 days ago • 2.82k • 63

Aleph-Alpha/Kolibri-1

Text Generation • 78B • Updated about 7 hours ago • 19.9k • 959

New activity in Aleph-Alpha/Kolibri-1 11 days ago

[Evaluation Report] Kolibri-1 FP8 on one RTX PRO 6000 (96 GB): serving numbers, and BEAM 100K from its own 262K window vs a 7K-token memory page

#10 opened 11 days ago by