The price gap between open-weight and closed frontier models: • GLM-5.3-Flash: ~$0.15 / $0.50 per million tokens (in/out) • Opus 5: $5 / $25 • GPT-5.6 Sol: $4 / $20 (current promo price, was $5/$30) Super cheap for a model that: – Nearly matches Opus 4.8 on Terminal-Bench (84.3 vs 85.0) – Beats it on DeepSWE and AutomationBench – 1M context, natively multimodal, only 18B active params of 320B total Why are you still paying Anthropic and OpenAI?