- December 2, 2025
- 2 min read
Enterprise Features Now Available on Friendli Dedicated Endpoints (Basic Plan)
- FriendliAI’s basic plan now includes features formerly reserved for the Enterprise tier, such as real-time metrics and logs.
- Users can now customize request-based autoscaling and host hundreds of LoRA adapters on a single endpoint with zero downtime.
- All workloads benefit from SOC2-compliant infrastructure, providing high-level security to teams of all sizes at no extra cost.
At FriendliAI, our mission is to make high-performance AI inference effortless, scalable, and cost-efficient for every team. As a major step toward this, Friendli Dedicated Endpoints now make Enterprise-level features available to all Basic plan users.
Starting today, several capabilities previously exclusive to the Enterprise plan are now included in the Basic plan, such as full access to metrics and logs, request-count-based autoscaling, and support for attaching hundreds of LoRA adapters to each endpoint. These upgrades give smaller teams access to the same production-grade tooling previously reserved for large-scale deployments.
What’s New?
- Observability with real-time performance metrics
- Customizable request-count-based autoscaling
- Request and response content logging (optional, opt-in)
- Support for hundreds of LoRA adapters
- SOC2 compliant infrastructure
Why Do These Changes Matter?
These upgrades give teams deeper visibility, smarter scaling, stronger security, and more flexibility when deploying models. Real-time metrics and logs give teams immediate visibility into system behavior, making it far easier to understand and resolve issues quickly. With request-count–based autoscaling, teams set their own queue thresholds, giving them fine-grained control over when and how their endpoints scale. Support for hundreds of LoRA adapters gives teams the flexibility to run more specialized and fine-tuned models on a single endpoint. And with SOC2-compliant infrastructure, teams can deploy these workloads with confidence, meeting modern security and governance standards.
Taken together, these upgrades make the Basic plan significantly more capable, giving teams access to the advanced tooling and operational controls used in large-scale deployments.
What Can Basic Users Do Now That They Couldn’t Before?
With this update, Basic users gain access to a set of capabilities that were previously reserved for Enterprise deployments. These features give teams far more operational control, observability, and scalability when running production-grade AI workloads.
1. View Full Metrics and Request Activity
| Previously | Now |
|---|---|
| • Basic users had no visibility into these operational insights. | • Monitor real-time throughput, latency, token processed, and replica counts over time
• Review request activity and troubleshoot issues more quickly |
2. Adjust Autoscaling Control Parameters
| Previously | Now |
|---|---|
| • Autoscaling could not be customized | • Users can now choose to adjust scaling based on how many requests are queued. |
3. Support for many LoRA Adapters
| Previously | Now |
|---|---|
| • Endpoints allowed only one LoRA adapter, which had to be fixed at creation time | • Run hundreds of LoRA adapters on a single endpoint
• Add or remove adapters in real time with zero service disruption |
4. Enable Request and Response Content Logging
| Previously | Now |
|---|---|
| • This capability was not available on the Basic plan | • View specific request and response content (when explicitly enabled)
• Get a clearer view of how the model is behaving • Spot and investigate requests that may require attention |
Start Exploring the New Basic Plan Today
This upgrade makes it easier to:
- Gain real-time observability with detailed performance metrics
- Run a large number of LoRA adapters on a single endpoint
- Exercise finer control over scaling
- Understand how the model is behaving through request and response content
If you’re already on the Basic plan, these features are ready for you to explore.
👉 Check out the full breakdown of updated Basic plan capabilities: https://friendli.ai/pricing.
Whether you're running large-scale inference for production AI products or exploring new models and configurations, these upgrades give you more capability at no additional cost. Experience FriendliAI’s enterprise-grade AI inference infrastructure, now expanded with Enterprise-level features accessible to Basic plan users.
Written by
FriendliAI Tech & Research
Share
General FAQ
What is FriendliAI?
FriendliAI is the Frontier Inference Cloud for Agents, delivering high throughput, low latency, and reliability at scale for agentic workloads. Through vertically optimized inference infrastructure, it delivers 2–5× faster output token speed and a 99.99% uptime SLA for high-volume production traffic.
How does FriendliAI reduce inference costs?
FriendliAI reduces inference costs through higher GPU utilization and optimized inference performance. FriendliAI's patented continuous batching technique, along with quantization, speculative decoding, KV cache offloading, multi-LoRA serving, and autoscaling, helps you serve more tokens with fewer GPUs, lowering your infrastructure costs without sacrificing performance.
Why should I choose FriendliAI over other inference providers?
FriendliAI is built for production AI agents, combining speed, reliability, and efficiency at scale. It delivers low-latency streaming, reliable long-context inference, and robust tool calling without compromising stability. According to independent OpenRouter benchmarks, FriendliAI consistently ranks among the top providers for throughput, latency, and reliability across leading open-weight models. See why customers choose FriendliAI
Which open-weight models does FriendliAI support?
Run today’s frontier open-weight models—including GLM, MiniMax, Kimi, DeepSeek, Qwen, Gemma, and more—with a simple API call. FriendliAI Model API gives you instant access to the latest models with optimized inference performance for production workloads. Explore models and pricing
How do I get started?
Getting started takes just a few minutes. [1] Sign up for FriendliAI, [2] Generate your API key, and [3] Make your first inference request with frontier open-weight models.
Still have questions?
If you want a customized solution for that key issue that is slowing your growth, support@friendli.ai or click Talk to an engineer — our engineers (not a bot) will reply within one business day.
