claude-opus-4-6 vs claude-sonnet-4-6
claude-opus-4-6 vs claude-sonnet-4-6: $5/$25 vs $3/$15 per 1M, the same workload priced on both; claude-sonnet-4-6 is 40% cheaper per request.
| claude-opus-4-6 | claude-sonnet-4-6 | |
|---|---|---|
| Provider · tier | anthropic · frontier | anthropic · mid |
| Input | $5 | $3 |
| Cached input | $0.5 | $0.3 |
| Cache write | $6.25 | $3.75 |
| Output | $25 | $15 |
| Long-context tier | none | none |
| Per request (2,000 in / 300 out) | $0.0175 | $0.010499999999999999 |
| Per month at 100k requests | $1,750 | $1,050 |
| …with the static prefix cached | $1,750 | $645 |
| Cheaper sibling (cascade) | claude-sonnet-4-6 | claude-haiku-4-5 |
Through TokenZero
Point your SDK at the gateway and keep both keys; the cache formatter, response cache and advisory routing act on every request, and each one comes back with a signed receipt showing the list-price baseline, actual cost and the saving.
OPENAI_BASE_URL=https://tokenzero.sh/w/<key>/v1 ANTHROPIC_BASE_URL=https://tokenzero.sh/w/<key>
Questions
Which is cheaper, claude-opus-4-6 or claude-sonnet-4-6?
At list prices (table 2026-10-26) claude-sonnet-4-6 is 40% cheaper per request on a 2,000-in / 300-out workload. Quality is a separate question: TokenZero's cascade only routes to a sibling it scores as sufficient and measures parity with shadow evaluation.
Where do these prices come from?
The providers' own pricing pages, re-verified on 2026-10-26; the gateway signs receipts with the same table.