Pricing & Capabilities lists every tracked model. Compare places three models side by side. Compliance & Residency cites HIPAA, FedRAMP, zero data retention, and EU residency for each cloud host. On Hardware, FP16, FP8, and INT4 are weights only, with a total when KV is computed. KV is 16-bit unless the card states another format.
Weekly on Mondays at 4:30 AM PT. Next refresh Mon 26 Oct.
Loading the index.
Index accurately reflects market data as of 26 Oct 2026.
This week
Scroll sideways to see more models.
Three models, side by side. An em dash means the vendor page does not publish that figure.
| Field | |||
|---|---|---|---|
| Context window | — | 1,000,000 tokens | 1,048,576 tokens |
| Input price per 1M | $10.00 | $10.00 | $2.00 |
| Output price per 1M | $50.00 | $50.00 | $10.00 |
| Cached input price | $1.00 | $0.25 | $0.10 |
| Text in / out | Yes / Yes | Yes / Yes | Yes / Yes |
| Image in / out | Yes / No | Yes / No | Yes / No |
| Audio in / out | No / No | No / No | Yes / No |
| Video in / out | No / No | No / No | Yes / No |
| Training cutoff | — | — | — |
$1,000.00
$1,000.00
$200.00
Lowest
Rises Dec 1, 2026 to $400 at 50M
Estimated cost weights each model's indexed input, cached-input, and output price per 1M tokens by the volume and the input/output split, and leaves a blank price as an em dash.
Estimate assumptions: 50M tokens, 75% input and 25% output, and 0% cached unless the cached share says otherwise.
