Compare
Add up to four models. The bill starts at 1M input tokens, 1M output tokens, and 1,000 units.
Monthly bill
Token models use the token counts. Other models use units times the flat rate.
Lowest bill: Mistral: Voxtral Small 24B 2507 · $0.40
Add another
Search by name, then add up to four.
| Mistral: Voxtral Small 24B 2507 | Anthropic: Claude Haiku Latest | |
|---|---|---|
| Provider | Mistral AI | Anthropic |
| Modality | text, audio | text, image |
| Context | 33K | 1M |
| Input | $0.1/1M | $0.1/1M |
| Output | $0.3/1M | $0.5/1M |
| Cache read | $0.01/1M | $0.01/1M |
| Cache write | Not published— | $0.125/1M |
| 1-hour cache write | Not published— | $0.2/1M |
Tiers and extras
- Mistral: Voxtral Small 24B 2507
- Audio: $100 per 1M tokens
- Anthropic: Claude Haiku Latest
- Web search: $0.01 per request; Above 100,000 prompt tokens: in $0.5 · out $2.5 · cache read $0.05 · cache write $0.625 · 1h cache write $1