Anthropic retired the Claude 3.x family and has now launched Claude 5 — the current line is Claude Opus 5, Claude Sonnet 5, and Claude Haiku 4.5. After running Claude across a few production contexts, here's where I've landed: the instruction-following is more reliable than the comparable GPT tier for anything that needs structured output, the context window is a full 1M tokens on Opus and Sonnet, and the prompt caching discount — 90% off cached tokens — is still the most aggressive in the industry and not talked about enough.
The big shift this generation is price: Sonnet 5 dropped to $2/$10 (down from Sonnet 4.6's $3/$15), so its output now undercuts GPT-5.6 Terra's $12/M. Opus 5's $25/M output is still real money on output-heavy workloads. Whether Claude ends up cheaper than the GPT tier for your use case depends mostly on one thing: how much repeated context you're sending. Here's the breakdown.
| Model | Input (per 1M tokens) | Output (per 1M tokens) | Context window |
|---|---|---|---|
| Claude Opus 5 | $5.00 | $25.00 | 1M |
| Claude Sonnet 5 | $2.00 | $10.00 | 1M |
| Claude Haiku 4.5 | $1.00 | $5.00 | 200K |
Source: platform.claude.com/docs pricing, September 2026. Claude Opus 4.x and Sonnet 4.x remain available at their prior pricing ($5/$25 and $3/$15); the 3.x family is retired. For the full cross-provider table, see the 2026 AI API pricing guide.
Claude Sonnet 5 is the model most teams should default to, now priced at $2.00/$10.00 per million tokens — level with GPT-5.6 Terra on input and cheaper on output. Where it consistently shines in practice: complex instruction-following, long-document reasoning, and careful adherence to structured output formats. Teams building on Sonnet tend to need fewer retries and less prompt engineering to get consistent structured outputs.
It includes the full 1M-token context window at standard pricing, so entire contracts, technical manuals, or codebases fit in a single request without chunking — and the price drop from the 4.x generation makes it materially cheaper to run at scale.
Claude Haiku 4.5 at $1.00/$5.00 per million tokens is where things get interesting. It's more expensive than the cheapest budget models (Gemini Flash-Lite, GPT-5.6 Luna) but delivers significantly higher quality — closer to Sonnet than the usual gap between a flagship and its mini tier.
If your workload needs better instruction-following or structured-output quality than a bottom-tier model delivers, but you can't justify full Sonnet pricing, Haiku 4.5 is worth evaluating seriously. It's especially strong at coding tasks relative to its price point.
Opus used to be a luxury — Claude 3 Opus launched at $15/$75 per million tokens. Opus 5 lands at $5/$25, holding the aggressive Opus 4.5–4.8 price point that makes top-tier reasoning practical. It's the one to reach for on genuinely hard reasoning, coding, and analysis; for the other 90% of production work, start with Sonnet 5 and only escalate when quality demands it.
This is Anthropic's biggest competitive pricing advantage and it's often overlooked. A cache hit costs just 10% of the standard input token rate — a 90% discount on cached tokens. (OpenAI and Google now price cache reads at roughly the same 10%, so the gap has narrowed, but Claude's caching remains best-in-class in practice.)
For applications with large, consistent system prompts or document contexts that repeat across requests, this is a massive lever. A 3,000-token system prompt on Sonnet 5 costs $0.20 per million tokens on cache hits vs $2.00 uncached.
Imagine you're building a RAG system that sends a 4,000-token document context plus a 500-token system prompt with each request. At 30,000 requests/month on Claude Sonnet 5 ($2/$10):
| Scenario | Input cost | Output cost (300 tokens avg) | Monthly total |
|---|---|---|---|
| No caching | $270.00 | $90.00 | $360.00 |
| With caching (system prompt + doc) | $27.00 | $90.00 | $117.00 |
That's roughly a 68% reduction just from caching, with no quality change. This is one of the most impactful optimizations available for high-volume Claude applications.
The headline comparison flipped this generation — Sonnet 5 at $2/$10 vs GPT-5.6 Terra at $2/$12 means Claude now matches on input and undercuts on output. Caching widens the gap further:
For workloads with minimal repeated context and very short prompts, GPT-5.6 Luna or a Gemini Flash tier is usually cheaper. For most production coding and analysis workloads, Sonnet 5 is now both strong and price-competitive.
Compare Claude 5 vs GPT-5.6 vs Gemini 3 for your exact token volumes and request rate.
Open the Calculator →