TileRT + GLM-5.3 on 8× AMD Instinct™ MI355X: #1 single-user performance on InferenceX AgentX, with ~2/3 of short-context speed retained at 1M tokens.
0 comments
No comments yet.
Related stories
- The Verge · 0 points · 2 days ago
- Ars Technica · 0 points · 1 day ago
- Hacker News · 1 points · 8 days ago
- DEV Community · 11 points · 13 days ago
- Hacker News · 7 points · 7 days ago
- DEV Community · 12 points · 14 days ago