Gemini Vision in Salesforce. Your GCP Controls Intact.
Vertex AI routing with IAM policies, VPC controls, and committed spend applied automatically. Multimodal from day 1.
66.5%of IT leaders report unexpected charges from consumption-based AI pricing models (Zylo, 2025)
GCP Already Solved This. New AI Vendors Undo It.
Your IAM policies, VPC controls, and committed spend don't apply to external AI vendors.
Another security review
Your GCP infrastructure already passed review. A new AI vendor means another round of questionnaires, DPAs, and compliance certs. Teams wait while the backlog grows.
“I don't want to send accounts to a separate system”
CTO, Financial Services
Secure this with Zero-Trust ArchitectureAI costs scattered across invoices
Your GCP committed spend covers Vertex AI. External AI platforms introduce separate billing, separate budgets, separate cost controls. Finance loses visibility.
“Why should that cost you a couple of million dollars?”
IT Leader, Manufacturing/CPQ
Secure this with Data MaskingExternal tools ignore GCP controls
Your IAM policies, VPC Service Controls, and org policies don't apply to external AI vendors. Data flows outside your security perimeter to infrastructure you don't control.
“Using our own data from our own Salesforce - that's gonna be a lot more useful to our sales reps”
VP of IT, Fortune 500 Insurance
Connect your model via Bring Your Own ModelVertex AI Integration, Your GCP Project Controls
Service Account Authentication
GPTfy authenticates to Vertex AI using your GCP service account stored in a Salesforce Named Credential. Your existing IAM roles govern which Gemini models are accessible. See how GPTfy's zero-trust architecture secures every callout.
GCP Project-Level Billing
All Gemini usage bills to your existing GCP project. Use your negotiated pricing, committed use discounts, and budget alerts. GPTfy adds no per-token markup.
Data Security, GCP Controls + GPTfy Masking
PII Masking Before Data Reaches GCP
GPTfy's Security Layer masks PII (names, SSNs, emails, phone numbers) in Apex before any data reaches Vertex AI. This adds a security layer on top of your existing GCP data protection.
VPC Service Controls Compatible
If your GCP project uses VPC Service Controls or Private Google Access, GPTfy works within those boundaries. Data flows to your VPC-protected Vertex AI endpoint. Network security policies stay enforced.
Salesforce Use Cases, Powered by Gemini
Record-Level Intelligence with Gemini Models
GPTfy sends structured Salesforce record data to Gemini as context. Choose Gemini 2.0 Flash for high-volume tasks or Gemini 2.5 Pro for complex reasoning. The AI Connection record controls model selection per prompt template.
Multi-Modal Capabilities in Salesforce
Gemini's multi-modal capabilities let GPTfy send document attachments, images, or structured data alongside text prompts. Use file analysis to process contracts, product images, or combine visual and textual data in a single request.
Why Choose Gemini for Salesforce
GCP IAM Policies Apply
Your existing GCP IAM roles, service account permissions, and org policies govern Gemini access. No separate credential management outside your Google Cloud security perimeter.
PII Masking + VPC Controls
GPTfy masks PII in Salesforce before the callout, and your GCP VPC Service Controls protect the Vertex AI endpoint. Two independent security layers, both under your control.
Existing GCP Billing
All Gemini usage bills to your GCP project with your negotiated pricing. No per-token markup from GPTfy, no separate billing relationship, no surprise invoices.
Powerful Capabilities
GCP IAM Authentication
GPTfy authenticates to Vertex AI using your GCP service account stored in a Salesforce Named Credential. Your IAM roles and org policies govern which Gemini models are accessible.
PII Masking + VPC Controls
PII is masked inside your Salesforce org before the callout. Your GCP VPC Service Controls and Private Google Access boundaries remain enforced throughout the data flow.
AI Response Audit Trail
Every Gemini request is logged as an AI_Response__c record with the prompt, masked input, model output, and token count preserved for compliance.
Existing GCP Billing
Gemini usage bills to your GCP project with your negotiated enterprise pricing and committed use discounts. No separate billing relationship or per-token markup from GPTfy.
Key Takeaways
- GPTfy authenticates to Vertex AI using a GCP service account stored in a Salesforce Named Credential. IAM roles govern model access.
- All Gemini inference usage bills to your existing GCP project with your negotiated pricing; GPTfy adds no per-token markup.
- GPTfy's Security Layer masks PII in Apex before data reaches Vertex AI, adding a masking layer on top of your GCP controls.
- VPC Service Controls and Private Google Access boundaries remain enforced; callouts target your VPC-protected Vertex AI endpoint.
- Supports Gemini 2.5 Pro, 2.0 Flash, and 1.5 Pro; select the model per AI Connection record without code changes.
- Gemini's multi-modal capabilities let GPTfy send Salesforce file attachments alongside text prompts for document analysis.
Frequently Asked Questions
GPTfy uses a GCP service account stored in a Salesforce Named Credential. The service account's IAM roles determine which Vertex AI models and features are accessible. Authentication follows Google's standard OAuth 2.0 service account flow, managed by the Salesforce platform's Named Credential framework.
Yes. GPTfy's callout architecture sends data from Salesforce directly to your Vertex AI endpoint. If your GCP project uses VPC Service Controls or Private Google Access, the callout targets your VPC-protected endpoint. Your network security policies remain enforced throughout the data flow.
GPTfy supports all Gemini models available through the Vertex AI API, including Gemini 2.5 Pro, Gemini 2.0 Flash, and Gemini 1.5 Pro. You select the model on the AI Connection record and can switch models without code changes. Different prompts can use different models based on complexity, cost, and latency requirements.
All Gemini inference costs bill directly to your GCP project. GPTfy does not add per-token charges or require a separate billing relationship. You use your existing GCP enterprise pricing, committed use discounts, and budget alerts. GPTfy's AI Response records track token usage so you can attribute costs to specific Salesforce use cases.
Yes. Gemini's multi-modal capabilities allow GPTfy to include document attachments (PDFs, images) from Salesforce Files alongside text prompts. This enables use cases like contract analysis, product image categorization, and visual inspection reports directly from Salesforce records.
Only PII-masked data reaches Vertex AI. GPTfy's Security Layer masks names, SSNs, emails, and phone numbers in Apex before the callout. Your GCP service account credentials are stored in a Salesforce Named Credential, encrypted at rest. Every Gemini request is logged to an AI_Response__c record for audit trails and compliance.
Run Gemini From Your GCP Project
30-minute demo. Vertex AI service account setup, PII masking, and Salesforce intelligence powered by Gemini with GCP billing intact.
Explore More Features
OpenAI in Salesforce
GPT-5.2 with function calling for structured outputs.
Claude in Salesforce
200K context window for complete Account history analysis.
DeepSeek R1
Chain-of-thought reasoning at 5-10x lower cost.
File Analysis
Multi-modal document analysis powered by Gemini Vision.
Security Layer
PII masking, Named Credentials, and audit trails.
What Is BYOM?
Guide to Bring Your Own Model architecture for Salesforce AI.