Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Long Context on NIAH
Loading...
99.8
Accuracy
Llama3.1
74.008
80.704
87.4
94.096
Dec 31, 2025
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Llama3.1
Size=8B, Type=Base
2025.12
99.8
Gemma3
Size=4B, Type=Base
2025.12
99.5
Youtu-LLM
Size=2B, Type=Base
2025.12
98.8
Qwen3
Size=4B, Type=Base
2025.12
83
Qwen3
Size=1.7B, Type=Base
2025.12
79.8
SmolLM3
Size=3B, Type=Base
2025.12
75
Feedback
Search any
task
Search any
task
