Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Long Context Retrieval on NIAH-Multi
Loading...
100
Accuracy
Kimi-K2
85.024
88.912
92.8
96.688
Jan 6, 2026
Accuracy
Updated 4d ago
Evaluation Results
Method
Method
Links
Accuracy
Kimi-K2
Length=64K, #Activated...
2026.01
100
MiMo-V2-Flash
Length=64K, #Activated...
2026.01
99.9
Kimi-K2
Length=32K, #Activated...
2026.01
99.8
DeepSeek-V3.1
Length=32K, #Activated...
2026.01
99.7
Kimi-K2
Length=128K, #Activate...
2026.01
99.5
MiMo-V2-Flash
Length=32K, #Activated...
2026.01
99.3
MiMo-V2-Flash
Length=128K, #Activate...
2026.01
98.6
DeepSeek-V3.1
Length=64K, #Activated...
2026.01
98.6
DeepSeek-V3.1
Length=128K, #Activate...
2026.01
97.2
MiMo-V2-Flash
Length=256K, #Activate...
2026.01
96.7
DeepSeek-V3.2
Length=128K, #Activate...
2026.01
94.3
DeepSeek-V3.2
Length=64K, #Activated...
2026.01
85.9
DeepSeek-V3.2
Length=32K, #Activated...
2026.01
85.6
Feedback
Search any
task
Search any
task