Lithos-metal: Qwen3.8-27B on 40-core M5 Max at 168 tokens/s
Description and use
Lithos‑metal showcases Qwen3.8‑27B running on a 40‑core M5 Max at 168 tokens per second, making it attractive for prototypes and internal workflows. Adoption involves manually vetting dependencies, integrating via API, and monitoring performance. Production readiness is moderate: regular updates and documentation maintenance are needed.
Use in a workflow
Lithos-metal: Qwen3.8-27B on 40-core M5 Max at 168 tokens/s may be useful when its README and activity match a concrete workflow.
Limitations and review notes
Medium: useful for prototypes or internal workflows, with dependency and maintenance checks before production.
- Source: hackernews
- Record date: 2026-10-17T04:06:23.000Z