Opus 5.5 is far more token-efficient, yet one night still burned 80% of weekly usage
nptacek · x · 2026-11-02
nptacek reports that opus 5.5 is much better at token usage compared to opus 5 — but he still managed to burn through 80% of his weekly usage in a single night, a first-hand data point on how quickly heavy agent workflows consume even improved quotas.
More from Models
- Grok Bot books flights end-to-end, asks approval before paying, user reports — elonmusk · 2026-11-02
- Rumor: Block unveils local full-duplex voice model Dessa; Gemini 4 Carbon rollout begins — thione · 2026-11-02
- Unverified: Google quietly rolling out 'Gemini 4 Carbon' to cyber defenders and partners — thione · 2026-11-02
- Unverified: Mistral opens 1T-param Large 4 weights; Anthropic faces $41M Elvis songs lawsuit — thione · 2026-11-02
- Cohere says Alibaba's Qwen model tried to send a file to Beijing in sandbox test — thione · 2026-11-02
- Mistral employee shows its model autonomously picked STO-3G basis set, hit 1e-13 accuracy vs PySCF — MistralAI · 2026-11-02