Companies now use an average of 12 models per month, yet the biggest hidden variable isn't which model you pick—it's the inference provider serving it and whether your cache hit rate stays above 85%. Peter Walker from OpenRouter explains how the same open-weight model can have wildly different cost and performance depending on the provider's caching policy, latency tail, and whether a stealth provider is reselling your prompts. The conversation covers why agentic workloads broke cache economics in February 2026, how ZDR policies vary even among "zero data retention" claims, and why the frontier labs are now competing across price and intelligence spectrums simultaneously. For engineers choosing production models, this is a practical map of the dimensions that actually move the bill.

• The inference provider layer—not just the model layer—now has 200+ competitors, and provider choice directly determines effective cost, latency, and data sovereignty.
• Cache hit rate is the dominant cost lever for agentic workloads: a cache miss can cost 50-100x more per token, and median agentic cache rates have climbed from ~50% to ~85% in 2026.
• Most organisations use a 'hub and spoke' model architecture with one 'daily driver' taking the majority of tokens and spend, plus specialised models for departments like customer support or sales.
• Zero data retention claims vary in trustworthiness by jurisdiction and provider size; some stealth models on OpenRouter are free specifically because they are not ZDR.
• The gap between frontier and open-weight models is widening in raw capability but narrowing in perceived value, as daily tasks saturate below frontier intelligence.
• Reasoning-only models and jagged intelligence are driving demand for fast decision models like Jev and OpenAI's decision model for latency-critical paths.

• 00:00 Introduction
• 00:38 The New Complexity
• 04:07 Provider Layer Competition
• 06:09 Choosing a Provider
• 08:42 Data Retention Realities
• 12:39 Model Portfolios
• 15:23 Latency and Reasoning
• 18:29 Cache Economics
• 23:35 Cost-Performance Sweet Spot
• 26:24 What Comes Next
• 29:13 Open Weight vs Frontier
• 32:02 Specialisation and Product
• 34:14 Where to Follow Peter

• OpenRouter: https://openrouter.ai
• OpenRouter Data: https://openrouter.ai/data
• DeepSeek: https://www.deepseek.com
• Anthropic: https://www.anthropic.com
• OpenAI: https://openai.com
• Google AI: https://ai.google
• Cerebras: https://www.cerebras.net
• CoreWeave: https://www.coreweave.com
• ElevenLabs: https://elevenlabs.io
• Artificial Analysis: https://artificialanalysis.ai
• Groq: https://groq.com

Peter Walker is Head of Insights at OpenRouter, where he analyses global model usage patterns across the inference layer.

This episode is presented by SparkLaunch and Dam Secure.

• SparkLaunch (https://sparklaunch.io/wizardry): Applied Wizardry listeners get an extra 20% off.
• Dam Secure (https://damsecure.ai/wizardry): Applied Wizardry listeners get their first month free.

• Justin McArthur, Executive Producer and Video Editor: https://www.linkedin.com/in/justin-mcarthur/
• Steph Hefferan, Editor: https://www.linkedin.com/in/steph-heff/

Follow Applied Wizardry on its new feed to keep getting every episode.

Dave Slutzkin, co-founder of Antiburn, joins Yaniv to break down what actually drives token costs in agentic development and how to bring them down. They cover how caching differs between Anthropic and OpenAI, when to reach for which model and thinking mode, and why the habit of compaction matters more than any other optimisation. If you're spending real money on AI-assisted development, you'll come away with concrete tactics for cutting token burn without cutting capability.

Key takeaways

• Context window depth is the dominant driver of token cost; compacting actively at 200,000-300,000 tokens matters more than any other single action.
• Cache expiry behaviour differs materially between providers: Anthropic's cache expires at exactly 1 hour, OpenAI's Codex cache is best-effort and often drops in 5-20 minutes, but OpenAI subscriptions do not charge for cache writes while Anthropic subscriptions effectively do through quota.
• Using a frontier model with a simple prompt to delegate to cheaper subagents is a practical pattern that balances capability and cost without requiring complex harness engineering.
• High thinking modes often produce worse results than medium or low because models overthink, invent requirements, and fix unrelated bugs; the benchmark sweet spot is typically medium.
• Skills files, MCPs, auto-memories, and system prompts silently bloat context windows; regular clean-up every month or two reduces both cost and quality degradation.
• Fast mode costs 2.5x per token but can improve output quality through tighter interaction loops; it is best used deliberately rather than left on by default.

Chapters

• 00:00 Introduction
• 00:38 The Token Burn Problem
• 03:12 Three Stages of Adoption
• 07:03 Model Choices Multiplying
• 11:14 The Cardinal Mistake: Depth
• 18:13 Worst-Case Usage Patterns
• 22:13 Picking Models by Feel
• 30:23 Thinking Modes and Overthinking
• 35:23 Fast Mode Trade-offs
• 48:46 Caching: Anthropic vs OpenAI
• 52:29 Context Window Ceiling
• 56:32 Cleaning Up Context Bloat
• 56:46 Where to Find Dave

Links and tools mentioned

• Antiburn: https://antiburn.com
• Anthropic Claude: https://www.anthropic.com
• OpenAI Codex: https://openai.com/codex
• Harvey: https://www.harvey.ai
• Legora: https://legora.com

About Dave Slutzkin

Co-founder of Antiburn, a company that helps developers understand and optimise their token spend for AI-assisted coding.

This episode is presented by SparkLaunch and Dam Secure.

• SparkLaunch (https://sparklaunch.io/wizardry): Applied Wizardry listeners get an extra 20% off.
• Dam Secure (https://damsecure.ai/wizardry): Applied Wizardry listeners get their first month free.

Credits

• Justin McArthur, Executive Producer and Video Editor: https://www.linkedin.com/in/justin-mcarthur/
• Steph Hefferan, Editor: https://www.linkedin.com/in/steph-heff/