Claude Sonnet 5 vs Claude Sonnet 4.6: is the upgrade actually cheaper?
By TechCompare · Updated · How we check
Sonnet 5 is both cheaper and smarter, one of the rare upgrades that needs no cost justification. Sonnet 4.6 only makes sense if you're locked into a deployment that specifically requires the older model's exact responses.
Anthropic: Claude Sonnet 5
Anthropic's own price
- Input
- $2
- Output
- $10
- Cached
- $0.20
per 1M tokens
Price checked 2026-11-12
Anthropic: Claude Sonnet 4.6
Anthropic's own price
- Input
- $3
- Output
- $15
- Cached
- $0.30
per 1M tokens
Price checked 2026-11-12
Cost Comparison
Based on 100,000 input tokens (50% cached), 5,000 output tokens, and 100 requests.Prices come live from OpenRouter at each maker's own price where it's listed. Temporary deals are left out.
Sonnet is Anthropic's coding-and-agentic workhorse tier. The 5 generation succeeded Sonnet 4.6 in mid-2026, bringing better capability and a price reduction of about 33% on input and 33% on output. For most Sonnet 4.6 customers, upgrading is a no-brainer purely on cost.
Same lineage, two generations. Compare list rates, cache reads, and what the newer model adds.
Side-by-side specs
| Spec | Claude Sonnet 5 | Claude Sonnet 4.6 |
|---|---|---|
| Input Cost (per M) | $2.00 (better on this spec) | $3.00 |
| Output Cost (per M) | $10.00 (better on this spec) | $15.00 |
| Cached Input (per M)Cached input is text the provider has already processed. It costs less than fresh input. | $0.20 (better on this spec) | $0.30 |
| Batch input / output (per M) | $1.00 / $5.00 | $1.50 / $7.50 |
| Context Window | 1M | 1M |
How they differ
Claude Sonnet 5 is priced at $2.00 per million input tokens and $10.00 per million output tokens, with a 90% caching discount ($0.20 per million) and a 50% batch discount ($1.00 per million inputs, $5.00 per million outputs). Claude Sonnet 4.6 is priced at $3.00 per million input tokens and $15.00 per million output tokens, with the same 90% caching discount ($0.30 per million) and the same batch discount.
For a typical 100K input + 10K output request, Sonnet 5 costs $0.30 vs Sonnet 4.6's $0.45, so the newer model is cheaper per call. The cache read cost drops too: $0.20 vs $0.30 per million. Both share the 1M context window and Anthropic's tool-use stack.
Verdict
Sonnet 5 undercuts Sonnet 4.6 on both base rows: $2.00/M input against $3.00 and $10.00/M output against $15.00. Cached input mirrors the input gap at $0.20/M versus $0.30/M, batch takes the same share off both, and context ties at 1M. A representative 100K input + 10K output call costs $0.30 on Sonnet 5 against $0.45 on 4.6. The capability is newer too, so the upgrade is a clean win on cost and quality unless 4.6 is pinned by a regression fixture.
Which should you pick?
Choose Claude Sonnet 5
Every new deployment. Sonnet 5 matches or beats Sonnet 4.6 on capability at a discount of about 33% on input, 33% on output and 33% on cached input.
Choose Claude Sonnet 4.6
Only if your code is pinned to Claude Sonnet 4.6's exact responses (regression tests, snapshot fixtures, reproducibility requirements). Otherwise migrate.