Dev Shares Token Split: Open Models Outused Closed 10:1 for Coding and Research
xeophon · x · 2026-09-14
Developer xeophon shared last week's token usage across coding, research and data analysis: open models outweighed closed ones roughly 10:1. It's an individual data point, but a notable signal that open-weight models can already carry most of a working developer's load.
More from Models
- CompBio Researcher Says a Qwen3.8-27B Fine-tune Beats Other Builds on Tool Calling, Cancels Claude — Usual-Carrot6352 · 2026-09-14
- Users say mystery model 'instinct' outperforms Grok and Muse in hands-on tests — Scobleizer · 2026-09-14
- Kokoro TTS ported to Apple Core AI: 54 voices running fully on-device with zero API cost — amos_gyamfi · 2026-09-14
- Multi-level simulation evals may be uninformative after Anthropic's Hacker Opus post — herbiebradley · 2026-09-14
- "Everyone is cheating on AI benchmarks": a hacker's plea to optimize for the real world — hackgoofer · 2026-09-14
- AI News Roundup: Opus 5.2 Leaks, GPT-6 Sol Previews, DeepSeek Code 2.0 Rumors, and the Frontier Slowdown Debate — WorldofAI · 2026-09-14