Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Time-to-first-token (TTFT) on Llama-3.1-8B-Instruct 128K input length
Loading...
14.5
TTFT (s)
TriangleMix + MInference
13.7
19.1
24.5
29.9
Jul 29, 2025
TTFT (s)
Updated 1mo ago
Evaluation Results
Method
Method
Links
TTFT (s)
TriangleMix + MInference
2025.07
14.5
TriangleMix + FlexPrefill
2025.07
16.1
FlexPrefill
2025.07
17.8
TriangleMix + XAttention
2025.07
17.8
MInference
2025.07
18
XAttention
2025.07
23.2
TriangleMix
2025.07
23.4
DuoAttention
2025.07
23.7
Dense
2025.07
34.5
Feedback
Search any
task
Search any
task