Product
Portkey Gateway
by Portkey
- Product website
- https://portkey.ai/
- Documentation
- https://docs.portkey.ai/docs/api-reference/inference-api/introduction
- Pricing
- https://portkey.ai/pricing
- Categories
- LLM Router
Description
Gateway for routing LLM requests with load balancing and monitoring.
AI-native score
How ready this product is for agents to use, on a 0–100 scale across four weighted dimensions. Security is one of them, so a high score is not a substitute for your own security or vendor-risk review. How is this scored?
Unrated. Only 23% of the scored decision fields are researched, so the grade is held until more are confirmed. The breakdown below shows what's been scored.
| Component | Raw | Weight | Contribution |
|---|---|---|---|
| Integration | 1/10 | 35% | 3.5 |
| Trust | 0/6 | 30% | 0.0 |
| Portability | 4/4 | 20% | 20.0 |
| Security | 1/4 | 15% | 3.8 |
Signals that fired
- Integration: SDK / developer kit (+1)
- Portability: Standard protocol access (+1) ; No proprietary-SDK lock-in (+1) ; Self-hostable (+1) ; Open source (+1)
- Security: Access controls — RBAC / ABAC (+1)
Signals not researched
These gaps cover the fields used in the AI-native score.
- Integration surfaces: llms.txt, WebMCP, Headless signup, MCP servers, Agent2Agent (A2A) servers, Native connectors, API specs, Command-line interfaces, Agent skills
- Trust & data policy: Trust center, Multi-factor authentication, Zero data retention, No generative AI training, Hard budget caps, Sandbox environment, Data export, Data import, Open standards, Documented provider swap
- Security & compliance: Compliance certifications
Trust & data policy
- Standard protocol access
- Yes
- No proprietary-SDK lock-in
- Yes
- Self-hostable
- Yes
- Open source
- Yes
- Source code
- https://github.com/portkey-ai/rubeus
Not yet researched Trust center, Multi-factor authentication, Zero data retention, No generative AI training, Hard budget caps, Sandbox environment, Data export, Data import, Open standards, Documented provider swap
Security & compliance
Certifications and access controls this product reports on its own pages. We confirm the vendor lists each one; for certifications we don't audit the underlying report. Part of the 15% security score.
Access controls
Not yet researched Compliance certifications
Integration surfaces
How an agent connects to this product and finds its documentation. The largest dimension of the AI-native score (35%).
Developer kits
-
Portkey Python SDK
Languages: Python
Not yet researched llms.txt, WebMCP, Headless signup, MCP servers, Agent2Agent (A2A) servers, Native connectors, API specs, Command-line interfaces, Agent skills
Offers
Free Forever
Free
Free Forever plan.
No recorded price in this currency
For prototyping and testing or evaluating enterprise POCs; not suitable for production workloads. Includes 10,000 recorded logs per month, with 3-day log retention and 30-day metrics retention.
Free Forever.
Production
$49/mo
The published price is $49/month; additional requests are $9 per 100,000.
No recorded price in this currency
Includes 100,000 recorded logs per month, $9 overages per additional 100,000 requests, 30-day log retention, 90-day metrics retention, and role-based access control with service-account API keys.
The plan is listed at $49/month.
Enterprise
Contact sales
No recorded price in this currency
Enterprise plan includes 10 million+ recorded logs per month, custom log and metrics retention, RBAC, SSO, granular budgets and rate limits, private-cloud deployment, data export to data lakes, VPC hosting, data isolation, and listed SOC 2 Type 2, GDPR, and HIPAA compliance offering.
Reviews
No approved reviews yet.
Catalog activity
Track how this entry changes as proposals are opened and merged.
Related products
Products from other publishers that share a job-to-be-done or category, most-shared first.
-
LLM Router
-
LLM Router LLM Security & Guardrails
-
AI / Agent Observability LLM Evaluation LLM Router Prompt Management
Ready to decide?
Unrated · 23% researched. Grade withheld until more fields are verified.