Riptide benchmarks its GPU attention kernels against AMD's on MI350X
Start here
Riptide published measured benchmarks of its GPU attention kernels against AMD's own kernels on the AMD Instinct MI350X, reporting both wins and losses with 95% confidence intervals and float64 correctness checks. Its kernels reach 3.32× AMD's best at head dim 128 causal training with S=512, but fall to 0.82–0.84× at 16k tokens for head dim 128 bf16. The kernels are not open source; Riptide offers them to partners for evaluation and licensing.
Generated from reports · updated 2 h ago
Report timeline
Follow the reports to see the event from different sides.
- Hacker News · AIRiptide benchmarks its GPU attention kernels against AMD's own on MI350X, publishing every loss
Riptide published measured benchmarks of its GPU attention kernels against AMD's own kernels on the AMD Instinct MI350X, reporting both wins and losses with 95% confidence intervals and float64 correctness checks. Its kernels reach 3.32× AMD's best at head dim 128 causal training with S=512, but fall to 0.82–0.84× at 16k tokens for head dim 128 bf16. The kernels are not open source and are offered to partners for evaluation and licensing.
Heat for this event
- Current, comparable range
- 9
- Peak, comparable range
- 10November 20 05:00
- Change in 24 hours
- –
The trend compares the same participants over hours observed continuously, so its range can be narrower than the current heat count. Hover or tap the chart for hourly heat; the arrow keys move through it.