COMMITS
April 6, 2026
S
feat: Add M5 Max (128GB) benchmark results (#49)
Sam committed
March 8, 2026
T
Add contributors
Tristan Bilot committed
T
Add last missing M1 Ultra
Tristan Bilot committed
J
Add M1 Ultra (48GPU) benchmarks (#47)
Jeremy Hale committed
January 4, 2026
June 6, 2025
D
Update README.md with M4, M4 Pro, M4 Max (#43)
David Jiménez committed
April 10, 2025
A
Add scaled dot product attention (#42)
Auke Wiggers committed
April 9, 2025
T
add M3 Ultra
Tristan Bilot committed
Z
add m3 ultra 60 core (#40)
Zhihao Deng committed
A
Add layernorm benchmark (#41)
Auke Wiggers committed
November 12, 2024
A
added m4pro results (#38)
Alexander Ziskind committed
A
added m4 results (#39)
Alexander Ziskind committed
A
M4max results (#37)
Alexander Ziskind committed
September 1, 2024
P
Benchmarks for A100 80GB (#34)
Patrick Haller committed
May 5, 2024
A
M2 Pro benchmark update with mlx 0.12.2 and torch 2.1.2 (#30)
Arnab Kumar Roy committed
March 20, 2024
I
M2 Ultra benchmark updated to 0.7.0 (#28)
Ivan Fioravanti committed
March 6, 2024
T
Use addmm in linear operation (#27)
Tristan Bilot committed
March 4, 2024
T
Fix cuda benchmark images (#26)
Tristan Bilot committed
T
Update README.md
Tristan Bilot committed
T
Update README.md
Tristan Bilot committed
T
M1 pro mlx0.5.0 (#25)
Tristan Bilot committed
I
Speed Memory Tradeoff (#23)
Ian Myers committed
T
Merge pull request #24 from Dave2011/readme-mixup
Tristan Bilot committed
D
fixed mixup
David Forster committed
February 27, 2024
T
Merge pull request #22 from TristanBilot/update_buggy_cuda_bench
Tristan Bilot committed
T
change order of RTX4090 and V100
Tristan Bilot committed
T
replace old Tesla V100 bench by fixed version with synchronize()
Tristan Bilot committed
T
handle runtime errors in the benchmark
Tristan Bilot committed
February 26, 2024
T
fix MLX import in utils
Tristan Bilot committed