Why CEREBRO kept it

Model routing halves coding agent costs; token optimization

Backlinks

Appeared in 1 briefing

• 2026-10-10

Related

Shares tags: ai/llm-mechanics · cerebro/signal

• Heads up: 'fast mode' for Opus is not included in your Claude sub. It bills you when you use it. https://t.co/wZvIlGh1kdx · 2026-10-10
• Haiku 5.5 is especially good value on tasks under 100,000 tokens, which make up around 90% of requests to our previous Haiku model. https://t.co/Qu8BZlizajx · 2026-10-08
• RT @claudeai: One more thing: we’re halving the price of cache reads on Claude Sonnet 5.5, to $0.10 per million tokens. That makes Sonnet 5.5 around 20% cheaper to run on most long-running work.x · 2026-10-08
• For prompts under 100K tokens, it's $0.10 / $0.50 per million tokens with cache reads at $0.01. Above 100K tokens, $0.50 / $2.50 per million, with cache reads at $0.05.x · 2026-10-08
• RT @lydiahallie: If you're on API billing, you can set Haiku 5.5's autocompact window to 100K so you stay in the cheaper token pricing tier! It's saved per model so this only applies to Haiku (incl. sx · 2026-10-08
• Claude Haiku 5.5hackernews · 2026-10-08

Also from openai.com

• How Jump Trading is scaling quant research with ChatGPTrss · 2026-10-07
• Advancing computer use with Ironcladrss · 2026-10-07
• A model guide for the GPT-6 familyrss · 2026-10-03
• Disrupting a coordinated model-distillation campaignrss · 2026-10-01
• Are you a Codex Original?rss · 2026-09-29
• Parallel cut research time and cost in half with GPT‑6 Astrarss · 2026-09-25