Long-context Question Answering on LoCoMo (test) benchmark leaderboard · SOTA2 Research

Long-context Question Answering on LoCoMo (test)

72.3LLM Score (Temporal)TRACE

Evaluation Results

Method	Links
TRACE2026.07		72.3	55.9	42.5	60.1	11.6	8.9	67.1	40.7	33.6	75.8	56.2	51.2	72.5	50.5	43.5
TRACE2026.07		69.3	53.5	38.8	58.3	13.3	10.4	59.9	35.7	28	67.9	48.9	43.9	66.1	45.2	37.9
Zep2026.07		66	52.8	45	40.6	24.1	18.6	60.6	37.1	25.1	55.2	39.9	33.9	57.5	41.1	33.7
Nemori2026.07		62.1	52.2	43	50	28.4	20.7	69.3	40.8	28.6	73	49.1	40.1	68.6	47	37.4
Mem02026.07		57.7	50.1	38.2	45.1	22.2	19.9	59.7	36.4	25.4	62.9	39.8	32.1	60.1	40.2	31.4
Full-Context2026.07		52.4	34.3	28.2	47.2	21.3	17.2	77.5	44.6	32.3	88.4	65.8	59.3	76.4	52.6	45.2
Nemori2026.07		50.7	48.2	40.9	47.6	20.5	15.1	51	28.4	19	68.8	45.9	36.6	60.4	41.6	32.9
A-Mem2026.07		49.4	41.3	35.7	27	11.3	9.2	50.3	25.4	18.7	58.9	37.9	31.4	53.3	34.7	28.6
RAG2026.07		46.2	37	31.6	28.9	10.3	8.4	39.6	22.1	15.6	54.7	36.3	30.4	48.6	32.2	26.6
A-Mem2026.07		46.2	29.8	24.9	42	22.4	17.8	61	36.8	25.9	68.6	49.8	44.3	60.9	41.6	35.2
Full-Context2026.07		43.5	37.9	33	44.1	17	13.4	70.6	27.4	21.1	88.1	45	33.3	72.8	38.5	29.7
RAG2026.07		43.5	30.1	26	38.5	20.2	15.7	40	27	16.5	57.4	41.3	36.3	50.2	35	29.3
LangMem2026.07		35.1	35.6	27.3	44.7	17	14	41.1	24.3	17	51.9	29.6	24	46	29.1	22.8
Zep2026.07		33.1	39	34.3	34.4	18.8	15.5	48	31.4	21.3	43.6	32.4	27.6	41.6	32.7	27.1
Mem02026.07		28.8	36	33.2	44.4	23	18	49.7	32.3	22.8	62.4	39.4	31.3	51.9	36.4	29.3
LangMem2026.07		21.5	26.2	22.2	43.5	24.9	19.8	53.7	33.3	25.4	58.3	36.6	29.6	48.8	33.1	26.7