Local LLM tips: Run gpt-osx-20b or Qwen on Mac
JoshPurtell · x · 2026-09-02
Supplementing the learning path, this post suggests practicing by constructing evals, exploring chain of thought, RL, and system prompting. For local experiments on Mac, it recommends running models like gpt-osx-20b via Tinker or Qwen 2.5 0.8b.
Related event: How Veteran ML Engineers Can Transition to LLMs(4 posts)→
More from Infra
- Cloudflare Agents emit OpenTelemetry traces, route directly to Braintrust for evals — ritakozlov · 2026-09-02
- Rabbi's take on DC moratorium: Using bans as leverage for environmental and labor concessions — joshua_saxe · 2026-09-02
- Intel exec: AI era security requires silicon-level design, not afterthoughts — BenBajarin · 2026-09-02
- PyTorch 2.14 released with 2,995 commits from 487 contributors — PyTorch · 2026-09-02
- Exllamav3 benchmarks: 700tk/s on 8x3090 setup — Leflakk · 2026-09-02
- GLM-5.3 Model Gets GGUF Quantization Release for Edge Deployment — unsloth · 2026-09-02