✳flâneur — a map of the web's best reading
Opus 4.7's New Tokenizer: What It Actually Costs — OpenRouter Blog
openrouter.ai · 1,100 words · saved by 1 readers
Anthropic changed the tokenizer in Opus 4.7. We looked at usage that shifted from 4.6 to 4.7 to measure exactly how it affects costs.
Opus 4.7's New Tokenizer: What It Actually Costs Justin Summerville · 4/27/2026 We Used Our Own Tokenizer to Get a Comparable Baseline We Used Our Own Tokenizer to Get a Comparable Baseline Opus 4.7 Produces 32–45% More Native Tokens Caching Absorbs Most of the Token Inflation Completion Length in Opus 4.7 Diverges Based on the Prompt Size Actual Cost Impact Methodology Anthropic announced that Claude Opus 4.7 improves the model’s understanding of inputs with a new tokenizer. This means that while the model price hasn’t changed ($5/M input, $25/M output), the same inputs will cost more
Explore this link on the map →related reading
- AI Model & API Providers Analysis | Artificial Analysisartificialanalysis.ai
- State of AI 2025: 100T Token LLM Usage Study | OpenRouteropenrouter.ai
- GitHub - brexhq/prompt-engineering: Tips and tricks for working with Large Language Models like OpenAI's GPT-4. · GitHubgithub.com
- A Picture is Worth 170 Tokens: How Does GPT-4o Encode Images?oranlooney.com
- tokens are getting more expensive - by Ethan Dingethanding.substack.com
- Inference characteristics of Llama · Cursorcursor.com
- GitHub - ccusage/ccusage: npx ccusage · GitHubgithub.com
- LLM Inference Performance Engineering: Best Practices | Databricks Blogdatabricks.com
- Introducing Claude Opus 4.7 \ Anthropicanthropic.com
- GitHub - karpathy/nanochat: The best ChatGPT that $100 can buy. · GitHubgithub.com
- GPT-4o mini: advancing cost-efficient intelligence | OpenAIopenai.com
- Branches · HazyResearch/intelligence-per-watt · GitHubgithub.com