Three AI Modes
Sorzio supports 300+ models from 16+ providers across three purpose-built AI modes:
- Generative AI (1.65x provider tokens + $0.0035/row base fee): Best for cold email openers, summaries, and structured extraction across OpenAI, Anthropic, Google Gemini, Mistral, xAI, and more. Optional Self-Healing Output Verification checks each draft against your source data and instructions, automatically revising ungrounded claims once before writing.
- Live Web Search (5 providers): Ground every row in real-time web research using
perplexity/sonar(around $8.78 per 1,000 rows all-in) or native web search on OpenAI, Anthropic Claude, Google Gemini, and xAI Grok. - Decision Mode (
typesafe-ai/jev-1.14.0, $1.00 per 1,000 rows base fee): Non-generative calibrated probability scoring for ICP qualification, boolean gates, and multi-choice classification ($0.0010/row base fee + 1.25x input tokens, roughly $1.01 per 1,000 rows). Evaluates up to 12 criteria in one pass with calibrated confidence scores.
Writing Effective Prompts
The quality of your results depends entirely on your prompt. Proven techniques:
- Be specific about what you want.Bad: "Write an email." Good: "Write a 3-sentence cold email to a VP of Engineering. Reference their company's recent product launch. Ask about their CI/CD pipeline challenges. Sign off as Alex from DevTools Inc."
- Include the format you want.If you want a one-liner, say "Respond in a single sentence." If you want JSON, specify the structure.
- Give examples."Here is an example of good output: [example]. Generate something similar for the following prospect."
- Set constraints."Do not use exclamation marks. Do not mention competitors. Keep the response under 50 words."
- Reference columns with double curly braces.
{{Company Name}}will be replaced with the value from the Company Name column for each row.
Using System Prompts
The system prompt sets the AI's identity and rules for the entire batch. Enter your system prompt in the "System Prompt" field (click to expand it if hidden). Example: "You are a senior SDR at a B2B SaaS company specializing in supply chain software. You write brief, specific, personalized cold emails. You never use filler phrases. You always reference something concrete about the prospect." Every row in your batch will follow these rules.
Understanding Costs and Token Baselines
Generative AI pricing combines a $0.0035 per-row base fee ($3.50 per 1,000 rows) with 1.65x the provider's actual input and output token cost (on Pro and Scale BYOK, both the markup and base fee drop to $0). In the model picker, pre-flight estimates start from a unified 1,000 input / 150 output token baselineper row (around $3.90 per 1,000 rows on budget models like GPT-4o-mini) and adapt automatically to your workspace's real median token usage after your first runs. Final settlement always bills exact post-run token usage on succeeded rows only.