| Model | USD / SIU | Spread to index | SIU per $1 |
|---|---|---|---|
| deepseek-v3.2 | $0.0018 | -76% | 552.7 |
| llama-3.3-70b-deepinfra | $0.0007 | -91% | 1423.5 |
| mistral-small-3.2-24b-instruct | $0.0006 | -91% | 1541.1 |
| qwen-2.5-72b-instruct | $0.0025 | -66% | 400.9 |
| claude-sonnet-5 | $0.0232 | +214% | 43.1 |
| gpt-5.1 | $0.0090 | +22% | 110.6 |
| gemini-3.1-pro-preview | $0.0139 | +88% | 72.0 |
Not yet published — the floor needs a measured hardware-rental rate for the reference configuration, which this print does not yet include.
The print recomputed under alternative cache and batch-discount policies, so the effect of those policy choices is stated rather than hidden inside one number.
| Policy variant | Dated SIU | Delta | Applies to |
|---|---|---|---|
| cache-enabled: 40% of T2 input at 10% of input price | $0.0050 | -32.2% | deepseek-v3.2, llama-3.3-70b-deepinfra, mistral-small-3.2-24b-instruct, qwen-2.5-72b-instruct, claude-sonnet-5, gpt-5.1, gemini-3.1-pro-preview |
| batch-discount: 50% off input and output | $0.0037 | -50.0% | deepseek-v3.2, llama-3.3-70b-deepinfra, mistral-small-3.2-24b-instruct, qwen-2.5-72b-instruct, claude-sonnet-5, gpt-5.1, gemini-3.1-pro-preview |
Don't take this page's word for it — every value below is checkable independently.
publisher(), and confirm it matches the recovered signer shown above.postedAt(bytes32) with the body hash shown above and confirm it returns a non-zero timestamp.