StartupBengaluru, India · Est. 2022
Portkey
AI gateway with guardrails for production agents.
Portkey builds an AI gateway that routes across 1,600+ models with integrated guardrails — caching, fallbacks, load balancing, rate limiting and observability — the control plane between a harness and every model provider. Strong presence in the APAC market.
Pros & cons
Each point cites its sources and says what kind of evidence backs it. A second reviewer tried to refute every claim before it was kept.
Pros
- DocsThe MIT-licensed gateway handles retries, fallbacks, load balancing and 50+ guardrails, and Gateway 2.0 (a pre-release) merges Portkey's enterprise gateway into the open-source repo.[1]
- DocsLightweight: Portkey claims under 1ms of added latency and a 122kb footprint.[1]
- User reportsOne user moved from LiteLLM's proxy to Portkey's free plan and found it much faster.[1]
- DocsLow entry cost: the free Developer tier records 10k logs a month, and Production costs $49/month.[1]
Cons
- DocsPalo Alto Networks announced in April 2026 that it intends to acquire Portkey and make it the AI gateway for Prisma AIRS, so the roadmap may now favour enterprise security buyers.[1]
- MetadataThe public gateway repo's last push was 25 May 2026, and 2.0 is still a pre-release branch.[1]
- DocsThe Developer tier keeps logs for only 3 days and is labelled as not suitable for production.[1]
- DocsPortkey says the Production tier is not recommended for organisations that need custom security controls or data residency, which pushes them to Enterprise.[1]
- Best for
- Teams that want a fast LLM gateway they can self-host, with fallbacks, retries and guardrails across many providers.
- Watch out for
- Watch how the Palo Alto acquisition affects the open-source gateway's release pace and hosted pricing.
Assessed Oct 29, 2026 · adversarially reviewedSuggest a correction
People
Co-founder & CTO
Harness engineering
Co-founder & CEO
Harness engineering
On X
No posts from them in our X feed yet.
Neighbours in the directory
New York, NY · Est. 2021
Serverless cloud built for AI and agentic compute.
Modal lets teams run inference, fine-tuning and arbitrary agent workloads as serverless Python without managing infrastructure — a common substrate for harness backends at scale.
Seattle, WA · Est. 2019
Durable execution for long-running agent workflows.
Temporal's durable execution engine is widely adopted to make agent pipelines reliable: state survives crashes, retries are automatic, and workflows run for minutes or months. A key piece of serious harness infrastructure.
$430M total · $300M Series D (Feb 2026) at $5B valuation
San Francisco, CA · Est. 2021
Durable functions and workflows for AI agents.
Inngest runs agent logic as durable steps — queueing, retries, concurrency control and human-in-the-loop waits — without operating your own workflow infrastructure.
Connections
Relations from the Harness Hub knowledge graph, as charted in the State of Harness reports.
Similar entries · 1
- LiteLLMProject