Share your thoughts, 1 month free Claude Pro on usSee more

Long-Context Retrieval on RULER (4K to 128K Sweep)

96.2Retrieval Accuracy (8K)

MrRoPE-Pro

Updated 9d ago

Evaluation Results

Method	Links
MrRoPE-Pro 2026.01		96.2	-	94.2	94.3	91.3	86.6	-
YaRN 2026.01		95.5	-	92.1	92.7	89.5	79.9	-
Nemo. 3 Nano 2026.04		95.2	96.2	93.8	90.3	85.9	-	-
Dense Attention 2025.12		95.07	96.16	94.8	92.33	87.69	-	93.21
BLASST 2025.12		94.7	96.17	94.61	91.81	87.06	-	92.87
Prism 2026.02		94.47	95.28	92.48	87.67	82.59	72.75	87.54
XAttention 2025.12		94.47	96.37	94.48	91.91	85.01	-	92.44
MInference 2026.02		94.46	95.43	93.42	87.22	83.07	71.04	87.44
Full 2026.02		94.38	95.42	93.38	87.98	84.72	77.77	88.94
XAttention 2026.02		94.3	95.17	93.28	89.06	82.31	70.52	87.44
Qwen 3 2026.04		94.1	95.5	93.5	90.1	-	-	-
MInference 2025.12		94.06	96.54	91.37	85.79	83.03	-	84.15
FlexPrefill 2025.12		93.67	95.99	92.73	88.14	81.14	-	87.72
FlexPrefill 2026.02		92.44	93.8	93.28	87.92	84.74	72.41	87.43
MInference 2026.02		92.37	95.08	89.67	86.01	76.53	70.36	85
Full 2026.02		92.35	95.01	90.04	87.24	79.93	75.09	86.61
Falcon H1 2026.04		92	92	89.7	83	77.2	-	-
Olmo Hybrid 2026.04		91.4	92.7	90	86.4	85	-	-
Olmo 3 2026.04		91.2	94.9	84.1	78.3	67.8	-	-
XAttention 2026.02		91.03	94.55	87.91	84.37	77.73	72.01	84.6
Prism 2026.02		90.95	94.84	87.69	86.88	78.58	72.65	85.27
Full-context 2026.04		90.1	-	95	83.4	85.5	76.3	86
Kimi Linear 2026.04		89.4	88.5	94	90.2	90.5	-	-
FlexPrefill 2026.02		87.61	90.89	87.82	85.58	78.27	73.42	83.93
Nemotron-H 2026.04		82	84.1	-	-	-	-	-
StructKV 2026.04		81.3	-	82.5	81.8	81.5	73.6	80.1
FastKV 2026.04		77.8	-	77.3	77.2	77.4	68.2	75.6
SnapKV 2026.04		75.6	-	76.8	72.9	75	67.7	73.6
GemFilter 2026.04		69.7	-	68.2	70.4	69.8	69.8	69.6
PyramidInfer 2026.04		66.5	-	-	-	-	-	-
Falcon Mamba 2026.04		49	68.2	23.8	4.7	0	-	-
H2O 2026.04		27.1	-	-	-	-	-	-
xLSTM 2026.04		20	28.6	12.4	-	-	-	-
StreamingLLM 2026.04		15	-	21.5	24.4	16.9	15	18.6
Recurr. Gemma 2026.04		-	44.9	-	-	-	-	-