AI Chatbot with Google Vertex AI on WordPress
SleekAI talks to Vertex AI through Google's OpenAI-compatible chat completions endpoint, so the chatbot on your WordPress site can run on gemini-2.5-pro or gemini-2.5-flash inside your own GCP project. Bring the project ID and a service-account-backed key.
♾️ Lifetime License available
Use the GCP project your data already lives in
Vertex AI is the enterprise-grade home for Gemini inside Google Cloud. The models are the same gemini-2.5-pro, gemini-2.5-flash, and Gemini Code variants you would call through the public Generative Language API, but the request lives inside a GCP project, with regional residency, VPC Service Controls, and billing on the same account that already runs the rest of your Google Cloud stack.
Vertex exposes an OpenAI-compatible chat completions endpoint on a per-project URL of the form https://{region}-aiplatform.googleapis.com/v1beta1/projects/{project}/locations/{region}/endpoints/openapi. SleekAI's OpenAI-compatible provider drops straight into that pattern: paste the URL, an access token minted from a service account with the roles/aiplatform.user role, set the model to google/gemini-2.5-flash, and the chatbot is live.
The traffic flows from WordPress directly to your Vertex endpoint, with no Sleek-hosted relay in the path. Conversations land in wp_sleek_ai_chats with the model name and token counts recorded, which makes it easy to line up usage against your GCP billing report. For teams already operating on Google Cloud, that single audit trail is often the whole reason to consolidate the WordPress chatbot onto Vertex.
Workflow
Wire SleekAI to Vertex AI in four steps
Enable the API
Mint an access token
Configure the provider
Reconcile and tune
Try it now
Ask the Vertex AI demo bot
Comparison
Generic chatbot vs SleekAI for Vertex AI
Generic chatbot
- Hard-coded to public OpenAI or Anthropic endpoints
- No support for Google's per-project Vertex URL
- Cannot use service-account-minted access tokens
- Forces a Google AI Studio key when you have a GCP project
- Logs and GCP billing report cannot be reconciled
SleekAI chatbot
- Works with Vertex's OpenAI-compatible chat completions endpoint
- Supports gemini-2.5-pro, gemini-2.5-flash, and other Gemini models
- Honors your GCP project, region, and VPC Service Controls
- Bills against the existing Google Cloud account, no Sleek markup
- Logs Gemini model name per chat for billing reconciliation
Features
What SleekAI gives you for Google Vertex AI
Per-project endpoint
Paste the Vertex AI OpenAI-compatible URL scoped to your project and region. SleekAI signs each call with the access token in the API key field, so the chatbot lands inside the GCP project that already owns your data.
Service-account-friendly
Use a token minted from a service account with the aiplatform.user role. Rotate via gcloud or workload identity federation, paste the current value into SleekAI, and the chatbot keeps answering with no further changes.
Mix Gemini variants
Run gemini-2.5-flash for fast marketing chat and gemini-2.5-pro for deeper documentation answers, both pointing at the same Vertex endpoint with different model strings selected per bot through Multibot.
Use cases
Where Vertex plus SleekAI fits
Google Cloud enterprises
Companies already on a Google Cloud commitment add a WordPress chatbot without onboarding a separate AI vendor. Spend rolls up on the same invoice, governed by the same IAM and org policies.
Residency-bound workloads
EU, UK, or APAC sites point Vertex at europe-west4, europe-west1, or asia-southeast1 so visitor prompts and Gemini responses stay inside the geography their data agreements require.
Unified AI governance
Platform teams already monitoring Vertex usage in Cloud Monitoring extend the same dashboards to the WordPress site, with chat logs and Gemini model counters in a single GCP project.
The bigger picture
Why Vertex is the right home for GCP shops
Companies that built on Google Cloud already invested in IAM hierarchies, org policies, VPC Service Controls, billing accounts, and Cloud Monitoring dashboards. Bolting a separate AI vendor onto a marketing site means duplicating all of that for one feature. Vertex AI solves that by giving Gemini a home inside the same GCP project.
The chatbot on the WordPress site can run on gemini-2.5-pro without the security team writing a new third-party risk review and without finance opening a new vendor record, because Vertex is just another Google Cloud service governed by the controls you already operate. Most WordPress chatbot plugins ignore Vertex entirely and assume you will paste a Google AI Studio key or, worse, an OpenAI key. SleekAI takes a different approach: any OpenAI-compatible endpoint is a first-class provider, which means Vertex's per-project URL plugs in directly with a service-account-minted access token in the key field.
Prompts stay in the region you chose, Gemini answers route back through the same project, and the conversation log in wp_sleek_ai_chats can be reconciled against Cloud Billing on demand. The chatbot becomes a normal GCP workload instead of a shadow integration.
Questions
Common questions about SleekAI for Google Vertex AI
AI Studio (api key from aistudio.google.com) is the consumer-grade Gemini API meant for prototyping. Vertex AI is the enterprise-grade Gemini surface inside Google Cloud, with project-level billing, IAM, regional residency, and VPC Service Controls. SleekAI supports both, but Vertex is the right choice for production GCP shops.
Vertex's OpenAI-compatible endpoint follows the pattern https://{region}-aiplatform.googleapis.com/v1beta1/projects/{project}/locations/{region}/endpoints/openapi. Replace region with us-central1, europe-west4, etc., and project with your GCP project ID. SleekAI uses that base URL for all chat completion requests.
Create a service account with roles/aiplatform.user, mint a short-lived access token via gcloud auth print-access-token or via workload identity federation in your CI pipeline, and paste the token into SleekAI's API key field. Rotate on whatever schedule your security policy requires.
Any model Vertex exposes through the OpenAI-compatible endpoint. As of recent updates, that includes google/gemini-2.5-pro, google/gemini-2.5-flash, and related variants. Set the exact model string in the SleekAI bot configuration so the request routes correctly.
Yes. If your Vertex endpoint is inside a VPC Service Controls perimeter, the WordPress host needs to reach it from an allowed network. Most teams deploy the WP frontend or an API Gateway in the same VPC, or use Private Service Connect, so the request never crosses the perimeter.
Gemini usage shows up on the same Google Cloud invoice as the rest of your project, broken down by model in Billing reports. SleekAI never marks up tokens or charges per request. The plugin license is a one-time fee through SleekWP and unrelated to inference cost.
Yes. Multibot lets each chatbot pick its own provider config and model. A common pattern is gemini-2.5-flash on the homepage for speed and gemini-2.5-pro on the docs site for depth, both reaching the same Vertex project under different model strings.
Vertex exposes text-embedding-005, text-multilingual-embedding-002, and other embeddings models through the same OpenAI-compatible surface. Point SleekAI's embeddings provider at the same base URL with an embeddings model name to build a fully Vertex-backed retrieval pipeline.
Pricing
More than 1000+
happy customers
Explore our flexible licensing options tailored to your needs. Upgrade your license anytime to access more features, or opt for a lifetime license for ongoing value, including lifetime updates and lifetime support. Our hassle-free upgrade process ensures that our platform can grow with you, starting from whichever plan you choose.
Lifetime ♾️
Most popular
EUR
once
- Unlimited websites
- Lifetime updates
- Lifetime support
...or get the Bundle Deal
and save €250 🎁
The Bundle (unlimited sites)
Pay once, own it forever
Elevate your WordPress site with our exclusive plugin bundle that includes all of our premium plugins in one package. Enjoy lifetime updates and lifetime support. Save significantly compared to buying plugins individually.
What’s included
-
SleekAI
-
SleekByte
-
SleekMotion
-
SleekPixel
-
SleekRank
-
SleekView
€749
Continue to checkoutBrowse more
- Solopreneur Chatbot
- AI Chat Summarizer
- Chatbot With PII Redaction
- Knowledge Base Search Bots
- AI Sales Reps
- Chatbot with Rate Limiting
- Lead Capture Bots
- AI Onboarding Coaches
- Help Scout Handoff
- Chatbot Powered by Gemini
- Chatbot With Usage Quota
- Document-Trained Chatbot
- Menu Recommendation Bots
- Chatbot with MCP Support
- Chatbot With EU Data Residency
- Stem Cell Clinics
- EMDR therapists
- Art therapists
- Vascular Access Centers
- Regenerative Orthopedic Clinics
- Assisted Living Facilities
- hearing clinics
- Lactation consultants
- Hyperbaric Oxygen Clinics
- Cosmetic surgeons
- Stem Cell Therapy Clinics
- Speech Therapists
- Wound Healing Clinics
- Optometrists
- Thyroid Clinics