| Model | USD / SIU | Spread to index | SIU per $1 |
|---|---|---|---|
| deepseek-v3.2 | $0.0018 | -83% | 541.6 |
| llama-3.3-70b-deepinfra | $0.0007 | -93% | 1398.5 |
| mistral-small-3.2-24b-instruct | $0.0007 | -94% | 1502.3 |
| qwen-2.5-72b-instruct | $0.0025 | -77% | 394.1 |
| claude-sonnet-5 | $0.0237 | +117% | 42.2 |
| gpt-5.1 | $0.0092 | -16% | 108.8 |
| gemini-3.1-pro-preview | $0.0402 | +268% | 24.8 |
| grok-4.6 | $0.0173 | +58% | 57.9 |
| claude-haiku-4-5 | $0.0077 | -29% | 129.0 |
| gpt-5.4-mini | $0.0053 | -51% | 187.2 |
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 | $0.007872 | -27.9% | 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, grok-4.6, claude-haiku-4-5, gpt-5.4-mini |
| batch-discount: 50% off input and output | $0.005522 | -49.4% | 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, grok-4.6, claude-haiku-4-5, gpt-5.4-mini |
Dated SIU is an assessment of qualifying inference work under this basket version, model/provider configuration and quality gate — not a prediction of production agent costs, which vary with workflow design, tool reliability, retry policy and context handling.
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.