Free
0
- Real-time guardrails for hallucinations, toxicity, PII/PHI, prompt injection, and jailbreaks, powered by Fiddler Centor Models
Fiddler AI: LLM Gateways
Founded 2018|Free tier, usage-based (per trace), Enterprise custom
What is Fiddler AI?
Fiddler AI is an enterprise platform for observing, evaluating, and governing AI agents and predictive ML models. The company was founded in 2018 in Palo Alto and started with explainability for classical ML models, then added monitoring, and now positions the product as a control plane for AI agents.
The platform covers agentic observability, continuous evaluations, guardrails, ML monitoring, and governance reporting. Evaluations and guardrails run on Fiddler Centor Models, task-specific models that run inside the customer's environment rather than calling an external LLM for each check. Fiddler reports enforcement latency under 80ms.
Fiddler sells mainly to enterprises in regulated industries and supports SaaS, VPC, on-premises, and AWS GovCloud deployment. The free tier covers guardrails, observability is billed per trace on the Developer plan, and VPC or on-premises deployment requires Enterprise.
Key features
Core capabilities this platform advertises.
- Agentic observability across multi-step agent workflows
- Continuous evaluations from testing through production
- Inline guardrails for PII, PHI, prompt injection, and jailbreaks
- AI governance, risk, and compliance reporting
Strengths and tradeoffs
What this tool does well, and the limitations to keep in mind.
Pros
- Covers AI agents and predictive ML models in one platform
- Evaluations and guardrails run in-environment, with no external LLM API call per check
- Free tier includes real-time guardrails
- Supports SaaS, VPC, on-premises, and AWS GovCloud deployment
Cons
- The free tier covers guardrails only; observability starts on the paid Developer plan
Plans & pricing
What's included in each plan, and how the tiers compare.
Developer
$0.002
Per trace
- Everything in Free, plus unified observability for agentic and predictive systems, custom evaluators and bring-your-own judge, RBAC and SSO, SaaS deployment
Enterprise
Custom
Contact sales for a quote
Common use cases
Enterprises in regulated industries (financial services, healthcare, insurance, government) running AI agents and predictive ML models
- Governing AI agents in regulated industries
- Monitoring predictive ML models for drift and performance
- Redacting PII and PHI on agent requests and responses
Using Fiddler AI with Respan
Respan covers the tracing and evaluation work Fiddler does and adds the AI router in front of it, so every model call is routed, traced, priced, and scored in one place. A bad score links straight to the trace behind it, and teams can start free with self-serve signup.
- One API for 1,000+ models, with automatic fallbacks when a provider errors or rate-limits
- Every LLM call, tool run, and agent turn in one nested trace
- Evals on test sets before shipping and on live production traffic after
- Hard spend limits per key, per customer, or org-wide
- Prompt versions that ship without a code deploy
Route, observe, and evaluate every LLM call with Respan
10KFree traces/mo
1,000+Models
Best Fiddler AI alternatives & competitors
Top companies in LLM Gateways you can use instead of Fiddler AI.
Compare Fiddler AI
Side-by-side comparisons with other tools in this category.
Best integrations for Fiddler AI
Companies from adjacent layers in the AI stack that work well with Fiddler AI.
