Wafer Blog company blog
Performance per dollar is getting faster and cheaper
How we served GLM5.2 on AMD MI355X at 2626 tok/s/node and 213 tok/s single stream at over 2x lower cost than Blackwell.
- Points
- 358
- HN comments
- 136
- Submissions
- 1
- Cited by
- 0 commenters
Hardware & ElectronicsAI & Machine Learning2026
On Hacker News
| Submitted | Title on HN | Points | Comments |
|---|---|---|---|
| Jul 3, 2026 | Performance per dollar is getting faster and cheaper by latchkey | 358 | 136 |
