Hugging Face's logo

Spaces:
optimum
/
llm-perf-leaderboard
Running

App Files Community
33
llm-perf-leaderboard / app.py

Commit History

test new benchmarks
0321f62

BenchmarkBot commited on

remove memory for now because there are errors
531390e

BenchmarkBot commited on

added example config
483e3a1

BenchmarkBot commited on

fix bench_df modifications
e89d633

BenchmarkBot commited on

show quantized on plot
37be3a4

BenchmarkBot commited on

allow quantized models on plot
c3c27bd

BenchmarkBot commited on

added * to quantized models scores
b084ebf

BenchmarkBot commited on

reduce weight class and model type sizes
f41da95

BenchmarkBot commited on

put everything in about tab
c382b2a

BenchmarkBot commited on

fix order and size
9243e5d

BenchmarkBot commited on

sort by tradeoff but don't show it
5490c7c

BenchmarkBot commited on

experiments by model type and weight class
a894537

BenchmarkBot commited on

add some explanationes
67b4a03

BenchmarkBot commited on

added some html to tabs
527deaa

BenchmarkBot commited on

added multi hardware
c796580

BenchmarkBot commited on

switch to tradeoff distance
804d27e

BenchmarkBot commited on

remove normalization
d0251b9

BenchmarkBot commited on

added llm perf score
223c247

BenchmarkBot commited on

remove memroy from composite
44b43cb

BenchmarkBot commited on

added composite score sorting
f0729a8

BenchmarkBot commited on

support querying the plot from url
fbc8c87

BenchmarkBot commited on

allow slower models on the plot
18468e8

BenchmarkBot commited on

indicate multiple optimizations
9b1f5f8

BenchmarkBot commited on