JAM has higher query latency than one-shot AOT systems but higher LoCoMo F1. Its one-time workspace construction is faster than A-MEM and Mem0 and slower than MemoryOS and LightMem. Against MemAgent, JAM lowers online latency from 58.08 to 13.81 seconds per query (reported 76.2% reduction) while raising F1 from 46.13 to 52.09.

JAM has higher query latency than one-shot AOT systems but higher LoCoMo F1. Its one-time workspace construction is faster than A-MEM and Mem0 and slower than MemoryOS and LightMem. Against MemAgent, JAM lowers online latency from 58.08 to 13.81 seconds per query (reported 76.2% reduction) while raising F1 from 46.13 to 52.09. · CiteArk

Measurement results

0/18 assessed

0 supported0 disputed or conflicting18 not assessed

The complete result group stays with its claim. Assessed rows come first; their judgments do not replace the whole-claim conclusion. Expand a row for conditions, sources, and evidence history.

Each row is a measurement with its reported value, latest available observation, and cumulative evidence state.
Metric and conditions	Reported	Latest observation	Evidence judgment
method: A-MEM · benchmark: LoCoMo	200.42 s/workspace	—	Not assessed
method: A-MEM · benchmark: LoCoMo	0.48 s/query	—	Not assessed
method: A-MEM · benchmark: LoCoMo	41.17 F1 points (0–100)	—	Not assessed
method: Mem0 · benchmark: LoCoMo	90.67 s/workspace	—	Not assessed
method: Mem0 · benchmark: LoCoMo	0.2 s/query	—	Not assessed
method: Mem0 · benchmark: LoCoMo	32.1 F1 points (0–100)	—	Not assessed
method: MemoryOS · benchmark: LoCoMo	41.36 s/workspace	—	Not assessed
method: MemoryOS · benchmark: LoCoMo	0.52 s/query	—	Not assessed
method: MemoryOS · benchmark: LoCoMo	31.67 F1 points (0–100)	—	Not assessed
method: LightMem · benchmark: LoCoMo	6.82 s/workspace	—	Not assessed
method: LightMem · benchmark: LoCoMo	0.22 s/query	—	Not assessed
method: LightMem · benchmark: LoCoMo	32.97 F1 points (0–100)	—	Not assessed

Reported and observed measurements

Assessments (0)

Runs (0)

Measurement results