Built with Astro, hosted on Cloudflare.
claudereports.com is built with Astro, hosted on Cloudflare. It measures traffic with Cloudflare Web Analytics.
| Technology | Version | First seen |
|---|---|---|
| JavaScript Frameworks & Libraries1 | ||
| 7.0.7 | 9 Nov 2026 | |
| Frontend Build Tools & CSS1 | ||
| – | 9 Nov 2026 | |
| CDN & Hosting1 | ||
| – | 9 Nov 2026 | |
| Analytics & Tag Management1 | ||
| – | 9 Nov 2026 | |
| Accessibility1 | ||
| – | 9 Nov 2026 | |
| Fonts & Icon Fonts1 | ||
| – | 9 Nov 2026 | |
The crew hasn’t read this site’s colors yet. We read a palette when we capture a site’s screenshot, busiest sites first. Meanwhile, browse palettes from other sites.
tori.ns.cloudflare.comdave.ns.cloudflare.comv=DMARC1; p=quarantine; adkim=r; aspf=r; rua=mailto:dmarc_rua@onsecureserver.net;User-agent: * Allow: / User-agent: GPTBot Allow: / User-agent: ChatGPT-User Allow: / User-agent: Google-Extended Allow: / User-agent: PerplexityBot Allow: / User-agent: ClaudeBot Allow: / User-agent: anthropic-ai Allow: / User-agent: Applebot-Extended Allow: / # Structured facts for language models (https://llmstxt.org) Allow: /llms.txt Allow: /llms-full.txt Sitemap: https://claudereports.com/sitemap-index.xml
# Claude Reports > As of 19 September 2026, Anthropic lists Fable 5.1 at $10/$50 per MTok, Opus 5 at $5/$25, Sonnet 5 at $2/$10 (now standard), and Haiku 4.5 at $1/$5. Fable 5.1 vs Opus 5 scores in The Journal are vendor-reported, not independent measurements. Site: [https://claudereports.com](https://claudereports.com) Last baseline update: 2026-09-19 Independent publication; not affiliated with Anthropic. ## Signature - [The Journal](https://claudereports.com/journal/): As of 19 September 2026, Anthropic lists Claude Fable 5.1 at $10/$50 per million tokens, Opus 5 at $5/$25, Sonnet 5 at $2/$10 (now permanent; it did not rise on 1 September), and Haiku 4.5 at $1/$5. Fable 5.1 cache reads cost $0.25, or 2.5% of base input, against 10% on the other Claude models. The Fable 5.1 versus Opus 5 scores in this journal are Anthropic-reported from the 1 September announcement, not independent measurements. ## Current Claude lineup (as of 2026-09-19) - Claude Fable 5.1 (`claude-fable-5-1`): most capable generally available Claude model, launched 2026-09-01. $10/$50 per MTok, 1M context, 128K max output. Cache reads $0.25/MTok. - Claude Opus 5 (`claude-opus-5`): launched 2026-07-24 and recommended by Anthropic as the starting point for most workloads. $5/$25 per MTok, 1M context, 128K max output. - Claude Sonnet 5 (`claude-sonnet-5`): speed and intelligence balance. $2/$10 per MTok standard; the announced increase to $3/$15 on 2026-09-01 did not take effect. - Claude Haiku 4.5 (`claude-haiku-4-5`): fastest and cheapest. $1/$5 per MTok, 200K context, 64K max output. - Legacy, not current: Claude Fable 5, Claude Opus 4.8 and earlier Opus versions, Claude Sonnet 4.6 and 4.5. - Anthropic routing guidance: start with Opus 5, escalate to Fable 5.1 for demanding reasoning and long-horizon agentic work or when Opus 5 at higher effort still falls short. - Primary sources: [Fable 5.1 announcement](https://www.anthropic.com/claude-fable-and-mythos-5-1), [Opus 5](https://www.anthropic.com/news/claude-opus-5), [pricing](https://platform.claude.com/docs/en/about-claude/pricing), [choosing a model](https://platform.claude.com/docs/en/about-claude/models/choosing-a-model). ## Core pages - [Home](https://claudereports.com/) - [The Journal](https://claudereports.com/journal/): As of 19 September 2026, Anthropic lists Claude Fable 5.1 at $10/$50 per million tokens, Opus 5 at $5/$25, Sonnet 5 at $2/$10 (now permanent; it did not rise on 1 September), and Haiku 4.5 at $1/$5. Fable 5.1 cache reads cost $0.25, or 2.5% of base input, against 10% on the other Claude models. The Fable 5.1 versus Opus 5 scores in this journal are Anthropic-reported from the 1 September announcement, not independent measurements. - [Reports](https://claudereports.com/reports/) - [Free tools](https://claudereports.com/tools/) - [Methodology](https://claudereports.com/methodology/) - [Sources](https://claudereports.com/sources/) - [Freshness](https://claudereports.com/freshness/) - [About](https://claudereports.com/about/) ## Reports - [Claude Fable 5.1 vs Opus 5 vs Sonnet 5: Reading Anthropic's September 2026 Lineup](https://claudereports.com/reports/claude-fable-5-1-vs-opus-5-vs-sonnet-5/) - [Claude Sonnet 5 Pricing Became Standard: What $2/$10 Means for Production Routing](https://claudereports.com/reports/sonnet-5-pricing-standard/) - [Claude Text Watermarking: What It Is, What It Is Not, and What It Cannot Prove](https://claudereports.com/reports/claude-text-watermarking/) - [Anthropic's Cyber-Eval Incident Disclosure: What Was Reported and What It Means for Trust](https://claudereports.com/reports/alignment-cyber-eval-incidents/) - [Claude vs GPT Coding Benchmarks: What the 2026 Scores Actually Say](https://claudereports.com/reports/claude-vs-gpt-coding-benchmarks/) - [Terminal-Bench and Agentic Coding: Why the Harness Is Part of the Score](https://claudereports.com/reports/terminal-bench-agentic-coding/) - [SWE-Bench Pro vs Verified: Why Frontier Coding Scores Need a Harder Yardstick](https://claudereports.com/reports/swe-bench-pro-vs-verified/) - [Claude Cost, Context, and Latency: The September 2026 Model Tradeoff Report](https://claudereports.com/reports/cost-context-latency/) - [Reasoning and Science Benchmarks: Reading Claude Claims Beyond Coding](https://claudereports.com/reports/reasoning-science-benchmarks/) - [Multimodal and Document Analysis Benchmarks: Claude, GPT, and the Work Hidden Inside PDFs](https://claudereports.com/reports/multimodal-document-analysis/) ## Free tools - [Claude vs GPT Model Comparison Explorer](https://claudereports.com/tools/model-comparison/): Filter Claude, GPT, and benchmark-reference models by provider, access, price, context window, and published coding benchmark evidence. - [LLM Benchmark Decoder](https://claudereports.com/tools/benchmark-decoder/): Pick an AI benchmark and get a plain-English read on what it measures, where it saturates, and how to treat Claude or GPT vendor claims. - [Claude API Cost vs OpenAI Cost Comparator](https://claudereports.com/tools/cost-comparator/): Estimate monthly API spend for Claude and OpenAI models from input tokens, output tokens, request volume, cache reuse, and long-context assumptions. ## Key citation rules - Cite benchmark claims with date, model version, benchmark version, harness, effort setting, and source provenance. - Treat provider launch tables, benchmark-owner leaderboards, and agent-plus-model rows as different evidence types. - Anthropic's Fable 5.1 announcement table compares Fable 5.1, Fable 5, and Opus 5, and includes Anthropic-reported GPT-5.6 Sol figures only where Anthropic published them (not OSWorld 2.0 or Humanity's Last Exam). Vendor announcement scores are not independent measurements. - Do not compare scores across benchmark versions. Anthropic reports Terminal-Bench 4.0 and OSWorld 2.0; OpenAI reports Terminal-Bench 2.0 and OSWorld-Verified. - Use the tools only as browser-side estimators; private task suites should make procurement decisions. ## Editorial disclaimer Claude Reports is an independent publication. It is not affiliated with Anthropic, OpenAI, or any benchmark owner.