Savings estimate · prices checked 2026-11-03
Claude Sonnet 5.5 vs Levanto Sage 2 for classification calls
List prices
| Model | $ per M input | $ per M output | Source |
|---|---|---|---|
| Claude Sonnet 5.5 | $2 | $10 | Anthropic pricing |
| Claude Haiku 5.5 | $0.1 | $0.5 | Anthropic pricing |
| Levanto Sage 2 | $0.05 | $1 | Levanto list price |
Checked 2026-11-03. Same table the public prices endpoint serves. Prices change; the audit always uses the current table.
Cost per call: our assumptions
- 500 input tokens per call (a ~100-token message plus a 400-token prompt, the audit’s defaults).
- 5 visible output tokens (“yes”/“no” plus formatting).
- 320 reasoning tokens at Claude Sonnet 5.5’s API default effort (
high), billed as output. This is our estimate, not a provider figure. - Levanto Sage 2 charged the same 500 input tokens (conservative: it needs no instruction prompt) and 2 output tokens, since it returns a decision and a probability.
That gives an estimated $4.25 per 1,000 calls on Claude Sonnet 5.5, against $0.027 on Levanto Sage 2.
Effort level changes the bill
Reasoning tokens are billed at the output price, so the effort setting can change Claude Sonnet 5.5’s cost a lot. Estimated cost per 1,000 calls by effort level:
| Effort | Est. reasoning tokens | Per 1,000 calls |
|---|---|---|
| low | 40 | $1.45 |
| medium | 120 | $2.25 |
| high (API default) | 320 | $4.25 |
| xhigh | 800 | $9.05 |
| max | 1600 | $17.05 |
If a lower effort still gives the answers you need, that is a saving with no new vendor. Test it the same way.
Monthly estimate by volume
“Share kept” is a scenario, not a prediction: the share of calls where Levanto Sage 2 is confident enough to keep. The rest go to Claude Sonnet 5.5 too, so the hybrid pays for both on those calls.
| Calls / month | All on Claude Sonnet 5.5 | Hybrid, 50% kept | 70% kept | 90% kept |
|---|---|---|---|---|
| 10,000 | $42.50 | $21.52 | $13.02 | $4.52 |
| 100,000 | $425 | $215 | $130 | $45.20 |
| 1,000,000 | $4,250 | $2,152 | $1,302 | $452 |
Estimates at the assumptions above. A full switch to Levanto Sage 2 would cost $2.70 at 100,000 calls, but only makes sense if it agrees with Claude Sonnet 5.5 on nearly everything.
When the switch is worth it
The audit only recommends a hybrid when it is at least 30% cheaper than staying put. With these prices that needs Levanto Sage 2 to keep at least 31% of calls at your agreement target. Below that, the saving rarely pays for the change.
- Good fit: the same yes/no or labelling prompt runs thousands of times (lead qualification, ticket triage, spam or intent checks, field extraction).
- Keep Claude Sonnet 5.5: free-text answers, tool use, multi-step reasoning, or inputs that differ a lot call to call.
Check it on your own calls
Paste 10 to 25 inputs with the answers Claude Sonnet 5.5 already gave into the free cutover audit. It measures agreement, picks the confidence threshold that holds your target and projects the cost with your token counts. How it works · Choosing a threshold · Other model pairs · Claude Sonnet 5.5 extraction cost · Reasoning effort cost