Anthropic’s Claude Haiku 5.5 pairs a 1M context with a 5x cost surge at 100k tokens. Here is the worked math on subagent loops, cache reads, and compaction.
Prompt Caching & TCO
13 Articles in this Category
KV-cache discount mechanics, context window compaction ROI, enterprise AI FinOps, and inference budget forecasting.
In autonomous agentic coding, the most expensive mistake is assigning frontier reasoning models to operational plumbing. On…
GPT-6.1 Sol cache reads cost $0.10 per million tokens, but writes cost $2.50. Worked examples show when reuse lowers the complete request bill.
Tsinghua University’s ICLR 2026 Cache-to-Cache (C2C) neural fuser eliminates inter-agent text generation to accelerate multi-LLM inference by…
Benchmarking Apple M5 Max/Ultra Mac Studio (oMLX, Qwen 3.8 27B, Bonsai 2 27B, qwen-image-2.1) against Thunderobot’s Ryzen AI Max+ 395 120B MoE SSD laptop.
Claude Sonnet 5.5 leaks reveal Anthropic’s roadmap. See Theo’s GPT-6 Astra audit, the 4.4x token bloat trap, and Jev routing cutting agent costs by 80%.
Gemini 3.8 Flash takes on GPT-6 Luna. Compare 73.7% vs 66.6% DeepSWE, $0.10/M token pricing, TPU v6e vs Astra routing, and 2.8% deception rates.
OpenAI drops GPT-6 Sol at $2/M input with 50% fewer errors, hitting 68.8% on DeepSWE v1.1 to match Claude Fable 5 at 80% lower cost. Full tech breakdown.
OpenAI releases GPT-6 Luna at $0.10/M input and $0.50/M output, scoring 66.6% on DeepSWE to match Claude Opus 5 at 93% lower per-task operational cost.
Anthropic released Claude Opus 5.5 on September 22, 2026: 66.4% Terminal-Bench, 54.4% FrontierCode, matching Fable 5.1 at $4/$20 with 60% cheaper prompt caching.