Claude Sonnet 5 pricing in 2026
What one task costs on Claude Sonnet 5
Per-million-token prices are hard to feel, so here are typical token counts for common jobs priced with Claude Sonnet 5. Figures are USD at the rates above and exclude prompt caching and batch discounts.
| Task | Tokens (in / out) | Cost |
|---|---|---|
| Code review of a 500-line pull request | 60,000 / 4,000 | $0.160 |
| Debug one failing test | 40,000 / 6,000 | $0.140 |
| Generate a landing page in one shot | 3,000 / 30,000 | $0.306 |
| Draft a 1,200-word article | 2,000 / 2,500 | $0.029 |
| Answer a question from a 20-page document | 25,000 / 800 | $0.058 |
| Multi-step agent run (10 tool calls) | 250,000 / 25,000 | $0.750 |
| All of the above, once | $1.44 |
Cheaper at the same or better capability: GLM-5.3 Flash runs the same set of tasks for $0.091 instead of $1.44.
Claude Sonnet 5 by Anthropic costs $2 per 1M input tokens and $10 per 1M output tokens — about 125.0k tokens per $1. Prices last verified 2026-09-19.
The verdict
Anthropic's developer workhorse. A strong index of 38 at $2 / $10 — roughly 60% cheaper output than Opus 5 — makes it the sensible default for most teams that want Claude's coding and tool-use without the flagship price.
A concrete example: sending 100,000 input tokens and getting back 20,000 output tokens with Claude Sonnet 5 costs about $0.40 at current prices.
Who is Claude Sonnet 5 for?
Sonnet 5 is Anthropic's developer workhorse — an index of 38 at $2 / $10 per 1M, with the coding strength and tool-use that make Claude popular, minus the flagship price. It's the default Claude for most production teams.
One caveat: Sonnet's tokenizer is famously wasteful on code, so real-world bills run higher than the per-token price implies. It still usually beats Opus on value for everyday work, and it beats most rivals on coding quality for the price.
Strengths & weaknesses
- Strength: Strong coding at mid-tier price
- Weakness: Wasteful tokenizer — among the highest token usage on Cursorbench, so real cost is higher than the per-token price suggests.
Claude Sonnet 5 key facts
| Provider | Claude pricing |
|---|---|
| Input price | $2 / 1M tokens |
| Output price | $10 / 1M tokens |
| Tokens / $1 | 125.0k |
| Intelligence / $1 | 47,500 |
| Context window | 1.0M tokens |
| Tier | balanced |
| Speed | Average |
| Intelligence index | 38 / 100 |
| Released | 2026-06-30 |
| Last verified | 2026-09-19 |
What Claude Sonnet 5 delivers for your budget
For a coding workload, here is roughly what each monthly budget buys:
| Budget | Tokens / month | Months of workload |
|---|---|---|
| $20 | 5.2M | 8.0x |
| $50 | 13.0M | 20.0x |
| $100 | 26.0M | 40.0x |
Other Claude models
How Claude Sonnet 5 compares to its closest rivals
Rather than repeating the full catalog on every page, here are the three current models nearest to Claude Sonnet 5 on the intelligence index — and why you might still pick one instead:
- GPT-5.6 Luna (ChatGPT) — $0.20 input / $1.20 output per 1M tokens. Why you might pick it: ChatGPT quality at a budget price.
- Gemini 3.7 Flash (Gemini) — $0.75 input / $3.75 output per 1M tokens. Why you might pick it: Frontier-adjacent index at Flash price.
- DeepSeek V4 Pro (DeepSeek) — $0.66 input / $1.98 output per 1M tokens. Why you might pick it: Unbeatable frontier value.
Claude Sonnet 5 — frequently asked questions
How much does Claude Sonnet 5 cost?
Claude Sonnet 5 by Anthropic costs $2 per 1M input tokens and $10 per 1M output tokens.
How many tokens per dollar does Claude Sonnet 5 give you?
On a typical 1:3 input-to-output mix, Claude Sonnet 5 delivers around 125.0k tokens per dollar — an intelligence-per-dollar score of 47500.
What is Claude Sonnet 5's context window?
Claude Sonnet 5 supports a context window of 1.0M tokens.
Is Claude Sonnet 5 worth it in 2026?
Claude Sonnet 5 is Claude's balanced tier model. Use the AI Intelligence Calculator to see how its value compares to every other model at your budget.