AvailableTokligence Guard
Sensitive Data ProtectionChrome extension that auto-detects and redacts sensitive data before sending to LLMs. Also supports AI coding agents.
Complete AI InfrastructureAvailable
Protect. Route. Trade.
The open-source toolkit for the LLM era. Protect your privacy, unify your AI providers, and trade AI tokens on our marketplace.
Your AI, Your Control.
Built by engineers from Baidu, NetEase, Shopee, SCOR, and OKX.
AvailableEarly Access
Tokligence Gateway
LLM RoutingHigh-performance gateway with dual OpenAI/Anthropic protocol support. Switch providers without changing code.
Get StartedComing 2026
Token Marketplace
Token TradingBuy AI token throughput when you need more capacity, sell your unused quota. The AI token marketplace that connects buyers and sellers.
Join WaitlistTokligence Guard
The Firewall for AI โ Protect Before You Prompt
Stop Leaking Sensitive Data to LLMs
Every time you paste code, emails, or documents into ChatGPT, you risk exposing API keys, personal information, and company secrets. Guard stops that.
How It Works
1
You Type
Enter text with sensitive data into any LLM chat
2
Guard Intercepts
Automatically detects sensitive data before it leaves your browser
3
Safe Replacement
Replaces with reversible placeholders like [EMAIL_001]
4
Auto Restore
LLM responses are automatically restored with original values
See It In Action
When You Send
You type
Polish my resume intro: John Smith | john.smith@gmail.com | +1-555-123-4567 Senior developer with 8 years experience...
LLM sees
Polish my resume intro: [PERSON_001] | [EMAIL_001] | [PHONE_001] Senior developer with 8 years experience...
When LLM Responds
You see
Here's your polished intro: John Smith | john.smith@gmail.com | +1-555-123-4567 Results-driven senior developer with 8+ years of experience...
LLM responds
Here's your polished intro: [PERSON_001] | [EMAIL_001] | [PHONE_001] Results-driven senior developer with 8+ years of experience...
Integration
Two Ways to Protect Your Data
Choose the integration that fits your workflow. Browser Extension protects personal data (e.g., names, addresses, emails) in LLM chats. Coding Agent secures credentials (e.g., API keys, passwords, secrets) in code and agent workflows.
Browser Extension
For web-based AI chats. Install our Chrome extension and Guard automatically protects your conversations on ChatGPT, Claude, Gemini, and 25+ platforms.
1Install extension from Chrome Web Store
2Open any supported AI chat (ChatGPT, Claude, etc.)
3Guard automatically intercepts and redacts sensitive data
4Chat safely with full protection
Best for: Web users, LLM chat users (ChatGPT, Claude, Gemini, etc.)
Install ExtensionCoding Agent / API
For CLI-based AI tools. Integrate via Tokligence Gateway to protect Claude Code, Codex CLI, Gemini CLI, and custom API integrations.
1Deploy Tokligence Gateway (open source)
2Generate API Key from Guard dashboard
3Configure Gateway with your API Key
4All requests are scanned before reaching LLM
Best for: Developers, AI coding agents (Claude Code, Codex CLI, Gemini CLI, etc.)
View Integration GuideWorks With All Major LLM Platforms
Guard protects your data on 25+ AI chat platforms โ with more added regularly
International
ChatGPT
Claude
Gemini
Perplexity
DeepSeek
Groq
Together AI
Mistral
Cohere
Grok
Copilot
Meta AI
Poe
You.com
China
ๅ
ๅฎYuanbao
่ฑๅ
Doubao
ๆบ่ฐฑๆธ
่จChatGLM
ๆๅฟไธ่จERNIE Bot
ๆตท่บAIHailuo
Kimi
้ไนๅ้ฎQwen
You can also add custom domains in the extension settings.
Enable AI Adoption Without Privacy or Compliance Risk
Tokligence Guard gives you peace of mind when using public AI models by protecting sensitive data at the source. It enforces real-time, on-device redaction to prevent accidental exposure of personal data, credentials, and secrets before any information leaves the device. When enabled, optional advanced detection applies minimal, pre-filtered text to cloud-based checks to identify unintentionally shared items such as API keys or addressesโwithout exposing raw content. Tokligence Guard operates with zero logging and zero data retention by design.
Multiregional Architecture
Unified control plane with regionalized data plane for global compliance and high performance
Zero Data Retention: We do not store your data or logs. Only minimal metadata required for service functionality is retained.
Core Design Principle: Unified Control Plane + Regionalized Data Plane for Global Compliance and High Performance
Free vs Paid: What You Get
Key features at a glance โ Guard protects 160+ data types
Features
Free
Paid Plans
Privacy Modes
๐ No Login Mode: 100% Local, Zero Data Leaves Browser
โ
โ
With Account: Access Cloud AI Detection
โ
โ
Detection Coverage
Email Addresses
โ
โ
Credit Cards (Visa, MC, Amex, UnionPay)
โ
โ
API Keys & Secrets (50 providers)
โ
โ
API Keys & Secrets (200+ providers)
โ
โ
Custom Regex Patterns
โ
โ
Sensitive Data Detection (100+ languages)
โ
โ
AI Name Detection (100+ languages)
โ
โ
AI Address Detection (100+ languages)
โ
โ
Coding Agent Sensitive Data Redact
โ
โ
Privacy & Processing
Local-First Processing
โ
โ
Works Offline
โ
โ
Cloud AI Detection
โ
โ
Higher Rate Limits
โ
โ
Support
Documentation
โ
โ
Email Support
โ
โ
Priority Support
โ
โ
Choose Your Protection Level
Getting Started
Free
$0forever
Essential protection for everyday use
- PII detection
- API Keys & Credentials (200+ Providers)
- Financial Data
- Web3 & Crypto Wallets
- 1,000 credits/month
- โ 100 LLM Chat protection or 1,000 Code Scans
Power User
Starter
$5/month
Global coverage with AI-powered detection
- Everything in Free, plus:
- 5,000 credits/month
- โ 500 LLM Chat protection or 5,000 Code Scans
- 100+ languages support
- AI-powered detection
Most Popular
Pro
$9.9/month
For professionals who need more capacity
- Everything in Starter, plus:
- 20,000 credits/month
- โ 2,000 LLM Chat protection or 20,000 Code Scans
- Higher rate limits
- Priority support
For Teams
Team
$99.9/month
For teams with shared billing and management
- Everything in Pro, plus:
- 300,000 credits/month
- โ 30,000 LLM Chat protection or 300,000 Code Scans
- Up to 20 seats
- Shared billing
- Team management
Enterprise
Enterprise Deployment Options
For organizations with unified email domains (@company.com). Choose the deployment model that fits your security and compliance requirements.
VPC / Self-Hosted
Deploy on your infrastructure with Docker, Docker Compose, or Kubernetes Helm charts. Data never leaves your network. Supports air-gapped environments with break-glass emergency access.
- Docker & Helm Deploy
- Air-gapped Support
- Break-glass Access
- Custom Domain
- Full Data Sovereignty
Enterprise SaaS
$12/seat/monthMinimum 20 seats
For companies without a data center. Fully managed by Tokligence with enterprise-grade SLA. Protects browser extensions and coding agent integrations.
- SSO (OIDC/SAML 2.0)
- SCIM 2.0 Provisioning
- Browser + Coding Agent Protection
- Unlimited API Credits
- Zero Data Retention
SSO Integration
Compatible with all OIDC and SAML 2.0 identity providers. Native SCIM 2.0 user provisioning, JIT (Just-in-Time) user creation, and group-based role mapping.
Tokligence Gateway
One Gateway, All LLM Providers
Stop rewriting code when you switch AI providers. Gateway gives you unified access to OpenAI, Anthropic, and Google with intelligent routing and cost tracking.
Dual Protocol Native
OpenAI and Anthropic APIs running simultaneously. Zero adapter overhead, native performance.
Intelligent Routing
Auto, passthrough, or translation modes. Route by model, cost, or latency.
Advanced Tool Calling
Cross-protocol translation, loop detection, session management. Verified with Codex CLI & Claude Code.
Cost Tracking
Token-level ledger with detailed accounting. Know exactly what you spend.
Get Started in Seconds
pip install tokligenceToken Marketplace
The Future of AI Token Economics
Traditional gateways are one-way: you consume tokens. Tokligence Marketplace enables two-way trading: buy when you need LLM capacity, sell your unused token throughput.
Buy Tokens
Access global LLM capacity. Find the best prices across providers and regions.
Sell Throughput
Monetize your unused token capacity. Turn idle throughput into revenue.
Smart Matching
Our marketplace connects sellers with unused token quota to buyers who need more throughput. Lower your AI costs during peak demand.
Be the First to Know
Join the waitlist for early access to Token Marketplace.
We respect your privacy. Unsubscribe anytime.
Security & Trust
Your Data, Your Control
We built Tokligence with privacy and security as core principles. We do not collect, store, or log the content of your requests. Only minimal metadata required for service operation is retained.
๐ซ
Zero Data Retention
We do not store your prompts, responses, or any content you process. Only minimal metadata required for billing is retained.
๐
No LLM Training
Your data is never used to train any AI models. We are a routing and protection layer, not an AI provider.
๐
No Content Logging
We do not log the content of your requests or responses. Your sensitive information stays private.
๐๏ธ
Compliance-First Design
Our architecture is designed for regulatory compliance with data residency controls and audit capabilities.
Contact Sales
Ready to take control of your AI infrastructure?
Get in touch for enterprise inquiries, partnerships, or questions. We typically respond within 24 hours.
VPC DeploymentCustom IntegrationVolume PricingCompliance & AuditDedicated SupportSLA Guarantee
Take Control of Your AI Infrastructure
Protect your data with Guard. Route your LLMs with Gateway.