Home / Companies / MintMCP / Blog / Post Details
Content Deep Dive

Portkey Pricing Explained (2026) | Mintive Blog

Blog post from MintMCP

Post Details
Company
Date Published
Author
Mintive
Word Count
2,850
Company Posts That Month
50
Language
English
Hacker News Points
-
Post removed?
No
Summary

Portkey, now operating as Palo Alto Networks’ Prisma AIRS AI Gateway following a May 2026 acquisition, provides a unified endpoint for routing requests to more than 1,600 language models while offering retries, fallbacks, caching, guardrails, cost tracking, and observability. Its hosted plans range from a free Developer tier with 10,000 monthly recorded logs and three-day retention to a $49-per-month Production tier with 100,000 logs, 30-day retention, and $9 per additional 100,000 requests, while enterprise capabilities such as SSO, VPC hosting, HIPAA BAAs, and custom retention require quote-based pricing; model-provider token charges remain separate. The platform emphasizes easy adoption through OpenAI-compatible APIs, configurable routing, semantic caching, virtual keys, and integrations with common AI development frameworks, while its open-source gateway can be self-hosted. The discussion notes that organizations with regulated or large-scale agent deployments may require longer audit retention, granular tool permissions, SSO and SCIM provisioning, and visibility into actions beyond model requests, presenting Mintive as an alternative focused on agent identities, governed MCP tool access, centralized credentials, and broader activity monitoring.

Trends Found in this Post
Trend Post Mentions Total Month Mentions Posts Companies MoM
LLM 16 3,838 757 183 -36%
MCP 15 7,233 710 184 -32%
Observability 11 1,924 468 147 -30%
AI Agents 8 3,889 962 215 -30%
AI Coding Assistant 3 1,245 338 108 -21%
Real-time 2 2,457 685 192 -49%
Secrets Management 2 1,735 373 111 -15%
AI Guardrails 1 277 127 44 -39%
Use This Data

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.