Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Downstream Task Evaluation on Downstream
Loading...
32.43
Throughput (tokens/s)
FairyFuse
7.2724
13.8037
20.335
26.8663
Apr 22, 2026
Throughput (tokens/s)
Memory Usage (GB)
Average Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Throughput (tokens/s)
Memory Usage (GB)
Average Accuracy
FairyFuse
Quantization=Ternary
2026.04
32.43
3.3
66
llama.cpp
Quantization=Q4_K_M
2026.04
26.15
4.1
65.1
llama.cpp
Quantization=Q2_K
2026.04
20.1
2.8
56.6
FP16
Precision=FP16
2026.04
8.24
13.5
67.3
Feedback
Search any
task
Search any
task
