Sorzio intelligence / 02
Explore 300+ models across 16+ providers, from generative work to live research and calibrated decisions. Choose the intelligence your workflow needs and understand the cost.
Model explorer / live catalog
Browse enabled models and see an illustrative batch estimate before you run.
Estimated Sorzio balance cost at the standard 1,000 input and 150 output tokens per row.
Loading the enabled model catalog...
Figures are estimates, not quotes. They use the public catalog and a 1,000 input / 150 output token baseline. Actual tokens, provider rates, search calls, and workspace settings affect cost. Per-run limits apply. Failed rows cost $0.00.
Provider marks
Marks identify providers; availability in the live catalog can change.
The choice
Choose Generative AI for drafting, transformation, or extracting structured information. Use Live Web Search when the answer needs current information from the web. Use Decision Mode to evaluate a defined rubric and route rows by calibrated probability before spending on downstream enrichment or personalized copy.
The table shows the pricing method for each mode. Model inference and search charges depend on actual usage, while the explorer uses an illustrative token baseline. A Pro or Scale workspace can use eligible AI keys with no Sorzio markup or AI base fee.
| Mode | Pricing method | Example or condition |
|---|---|---|
| Generative AI: platform balance | 1.65x provider token cost + $0.0035 per row | Depends on model and tokens |
| Generative AI: Pro or Scale BYOK | 0% Sorzio markup and $0 Sorzio base fee | Provider charges for actual usage |
| Live Web Search | 5 supported providers; native search fees at 1.65x | Perplexity Sonar approximately $8.78 per 1,000 all-in |
| Decision Mode | $1.00 per 1,000 rows base + 1.25x input tokens | Approximately $1.01 per 1,000 |
The details
On platform balance, Generative AI costs 1.65 times the provider token cost plus a $0.0035 base fee per succeeded row. Estimates start with 1,000 input and 150 output tokens per row; actual usage and workspace token profiles can differ.
Pro ($39/mo) and Scale support eligible AI provider keys with 0% Sorzio markup and no Sorzio AI base fee. You pay the provider directly for its usage. Model eligibility and the provider's own pricing still apply.
Live Web Search grounds AI output in current web information through supported providers. Decision Mode uses typesafe-ai/jev-1.14.0 to return calibrated probabilities for keep, reject, and unsure routing. Its $1.00 per 1,000 row base fee excludes input token costs.
No. Sorzio settles succeeded rows only, so failed rows cost $0.00. Costs shown in this explorer are estimates for successful rows and vary with the model, actual tokens, and any search calls.
Ready to model a complete pipeline? Explore Flow blueprints or compare plans on pricing.