Qwen 3.8 27B Scores 52, Matching Top Models Locally
Scobleizer · x · 2026-08-18
Qwen 3.8 27B achieved a score of 52 on Artificial Analysis, a massive milestone. It is the first time a model that can be run locally on a laptop is basically on par with the best proprietary models. This marks the point where local open-weight models are capable enough for most everyday tasks.
Related event: Qwen3.8-27B Benchmarks Rival Top Closed Models, Runs on a Single GPU(9 posts)→
More from Models
- Rumor: OpenAI to launch 'Astra' model this week — mark_k · 2026-08-18
- Qwen3.8-27B inference speed boosted to 62 tok/s — TheMoonMidas · 2026-08-18
- Running Qwen 27B at F16: Performance and VRAM Needs — Blues520 · 2026-08-18
- Qwen3.8-27B Benchmarks on M2 Ultra 192GB — planetearth80 · 2026-08-18
- Pangram's latest model enables 100% AI generation — soumitrashukla9 · 2026-08-18
- DeepSeek harness praised as visionary despite rough edges — aiamblichus · 2026-08-18