22 days ago
-
benchmarks/locomo-textv2-rq-p4-k32-seed0-20261004T015301Z: locomo s12500-k32-p4-rq
-
benchmarks/locomo-textv2-rq-p4-k32-seed0-20261004T014103Z: locomo s15252-k32-p4-rq
-
benchmarks/locomo-textv2-rq-p3-k32-seed0-20261004T012016Z: locomo base-k32-p3-rq
-
benchmarks/locomo-textv2-rq-p3-k32-seed0-20261004T012016Z: locomo s12500-k32-p3-rq
-
benchmarks/locomo-textv2-rq-seed0-20261004T010740Z: locomo s15252-k42-p1-rq
-
benchmarks/locomo-textv2-rq-seed0-20261004T010740Z: locomo s15252-k32-p1-rq
-
benchmarks/locomo-textv2-rq-seed0-20261004T010740Z: locomo s15252-k42-p3-rq
-
benchmarks/locomo-textv2-rq-seed0-20261004T010740Z: locomo s15252-k32-p3-rq
-
benchmarks/locomo-textv2-k32-p1-base-seed0-20261004T002130Z: locomo base checkpoint, text store, k=32, prompt p1
-
benchmarks/locomo-textv2-k42-p2-t2-seed0-20261004T001533Z: locomo step15252-prompt-only
-
benchmarks/locomo-textv2-k42-p2-t2-seed0-20261004T001533Z: locomo step15252
-
benchmarks/locomo-textv2-k42-p2-t2-seed0-20261004T001533Z: locomo base
-
benchmarks/locomo-textv2-k42-p2-t2-seed0-20261004T001533Z: locomo step12500
23 days ago
-
benchmarks/locomo-textv2-k42-seed0-20261003T234536Z: locomo base
-
benchmarks/locomo-textv2-k42-seed0-20261003T234536Z: locomo step15252
-
benchmarks/locomo-textv2-k42-seed0-20261003T234536Z: locomo step12500
-
benchmarks/locomo-textv2-k32-seed0-20261003T232337Z: locomo step12500
-
benchmarks/locomo-textv2-k32-seed0-20261003T232337Z: locomo step15252
-
Publish deep v3 fine-tune (22-source data, window-wide retrieval objective): update 12,500 recommended, final 15,252 and evidence alongside
29 days ago
-
benchmarks/k32-seed0-20260927T174023Z: summary after rerun
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-54-44Z-million-in-longmemeval-nmem_deep_k32-571ee6
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-54-44Z-million-in-locomo-nmem_deep_k32-7fa5d6
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-54-44Z-million-in-memoryagentbench-nmem_deep_k32_qwindow-d7b5c6
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-54-44Z-million-in-longmemeval-nmem_deep_k32-867839
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-54-44Z-million-in-memoryagentbench-nmem_deep_k32-157e64
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-54-44Z-million-in-memoryagentbench-nmem_deep_k32_qwindow-23ab0b
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-54-44Z-million-in-memoryagentbench-nmem_deep_k32_qwindow-29a513 json_answer_v1 rescore
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-54-44Z-million-in-memoryagentbench-nmem_deep_k32_qwindow-29a513 json_answer_v1 rescore
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-54-44Z-million-in-memoryagentbench-nmem_deep_k32_qwindow-29a513
-
benchmarks/k32-seed0-20260927T174023Z: summary
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T17-40-24Z-million-in-longmemeval-nmem_deep_k32-e5fc76
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-09-04Z-million-in-memoryagentbench-nmem_deep_k32-4c61fb
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-23-28Z-million-in-memoryagentbench-nmem_deep_k32-af4a84
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-10-14Z-million-in-memoryagentbench-nmem_deep_k32-dae403
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T17-40-24Z-million-in-memoryagentbench-nmem_deep_k32-7bc0e1
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-10-04Z-million-in-memoryagentbench-nmem_deep_k32-10274a
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T17-40-24Z-million-in-memoryagentbench-nmem_deep_k32-586e8f
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-02-48Z-million-in-memoryagentbench-nmem_deep_k32-5cb424
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-02-43Z-million-in-memoryagentbench-nmem_deep_k32-122b76
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T18-02-22Z-million-in-memoryagentbench-nmem_deep_k32-29c657
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T17-55-34Z-million-in-memoryagentbench-nmem_deep_k32-fe5e0e
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T17-55-23Z-million-in-memoryagentbench-nmem_deep_k32-231ec9
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T17-55-28Z-million-in-memoryagentbench-nmem_deep_k32-8c2f1d
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T17-48-06Z-million-in-memoryagentbench-nmem_deep_k32-1cc319
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T17-48-05Z-million-in-memoryagentbench-nmem_deep_k32-56192c
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T17-47-58Z-million-in-memoryagentbench-nmem_deep_k32-270701
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T17-40-24Z-million-in-memoryagentbench-nmem_deep_k32-d425f8
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T17-40-24Z-million-in-memoryagentbench-nmem_deep_k32-afad95
-
benchmarks/k32-seed0-20260927T174023Z: 2026-09-27T17-40-24Z-million-in-memoryagentbench-nmem_deep_k32-c86e65
-
benchmarks/k32-seed0-20260924T235851Z: completed tasks (sweep stopped early)