Running Kimi K3 on MI355X at Better Performance per Dollar Than B300

wafer.ai
August 2, 2026
1 min read
58/100
Summary
Wafer introduces Kimi K3, an open-source large language model (LLM) designed for enterprise use, emphasizing speed and efficiency. The model incorporates memory capabilities to enhance performance and user experience.
Coverage & discussions
Hacker News · wafer.ai
Community discussions
Historical count: 5 · 5 saved in sample · Historical sample
Analysis ready
Community Sentiment
Positives
- The MI355X is touted for its cost efficiency, with some commenters thrilled about possible better performance per dollar compared to the B300, shaking up the market.
- The excitement around using AI for setup and benchmarking suggests a future where automation can streamline complex processes, potentially democratizing access to advanced AI tools.
Concerns
- Criticism is flying over the misleading marketing, with commenters pointing out that the performance comparisons seem exaggerated and unfair, raising doubts about the integrity of the claims.
- There's a growing frustration about the lack of transparency regarding training data and methods, with many insisting that true 'open source' should include all components, not just weights.
Related Articles

Control the Ideas, Not the Code
Jul 13, 2026

I Am Not a Reverse Centaur
Jun 12, 2026

Qwen3.5 122B and 35B models offer Sonnet 4.5 performance on local computers
Feb 28, 2026

The reverts will continue until morale improves
Oct 19, 2026

Harness Engineering 101: The LLM is the brain, the harness is the body
Oct 19, 2026