Helicone vs Braintrust Gateway
The question that decides it: Do you want to measure production traffic after the fact, or gate quality before you ship?
Our verdict
Neither, if the gateway is the point — both are platform features with real uncertainty attached. If you must pick: Braintrust when your team already runs evals there and wants CI quality gates; Helicone only if you self-host it and want request analytics.
Why
Both of these are observability products with a gateway attached, not gateways with observability attached, and that framing matters. You would pick either because you already want the platform.
The two have different unresolved risks, and you should weigh both rather than treating one as safe. Helicone was acquired by Mintlify in March 2026 and is reported to be in maintenance mode with no new features planned; its standalone Rust gateway has not had a functional commit since July 2025. Braintrust's hosted Gateway is free during public preview with pricing to be announced before general availability — which means the cost of a component on your critical path is currently unknown.
On capability they are genuinely different products. Braintrust is eval-centric: spans, scorers, datasets and CI quality gates, so you can block a deploy when output quality regresses. Helicone is production-analytics-centric: sessions, per-user breakdowns, custom properties, an HQL query language, alerting. One tells you whether a change is safe to ship; the other tells you what is happening to real users now. Teams often want both, and neither replaces the other.
Two practical differentiators. Helicone is Apache-2.0 and genuinely self-hostable, which is a real hedge against the maintenance-mode risk — you can keep running it whatever Mintlify decides. Braintrust self-hosting is effectively Enterprise-only, so its preview-pricing risk has no equivalent escape hatch. Conversely, Braintrust supports unlimited users on all plans and offers time-limited temporary credentials for frontend and mobile clients, which is a genuinely useful feature neither of the others here has.
Neither offers real-time guardrails. Braintrust evaluates after the fact by design; Helicone's guardrails and PII redaction are weaker than Portkey's or OpenRouter's. If you need to block a bad prompt or response inline, look elsewhere.
Which one, concretely
Choose Helicone if
- You want production request analytics with a query language and alerting
- You want Apache-2.0 self-hosting as a hedge against vendor direction
- You need per-user and per-session cost attribution
- You want config-as-code support
Choose Braintrust Gateway if
- Your team already runs evals and tracing in Braintrust
- You want CI quality gates that block deploys on output regression
- You need unlimited users without per-seat cost
- You want temporary time-limited credentials for frontend or mobile clients
What catches people out
- Helicone: acquired by Mintlify March 2026, reported maintenance mode, no new features planned.
- Braintrust: hosted Gateway pricing after general availability is unannounced, so your future cost is unknown.
- Braintrust's jump from free Starter to $249/month Pro is steep, with per-GB and per-score overages that agent workloads can inflate quickly.
- Braintrust self-hosting is effectively Enterprise-only, unlike Helicone which is Apache-2.0.
- Neither offers inline guardrails. Braintrust is post-hoc by design; Helicone's are weak.
- Braintrust has a narrower provider list (18) and no public model-catalog endpoint.
Side by side
4 of 17 fields differ, marked with a dot. Every figure links to the vendor page it came from. Blank values read Not published rather than No — silence from a vendor is not a negative answer.
| Field | Helicone | Braintrust Gateway |
|---|---|---|
| Ease of leaving Derived score, higher is easier | 100/100 Easy to leave | 84/84 Some work to leave |
| What kind of product Category | Managed gateway | Managed gateway |
| Who runs it Deployment model | Managed or self-host | Managed or self-host |
| Licence Licence | Apache-2.0 | Open core |
| Models available Models available | ~100 | ~100 |
| Model providers reachable Upstream providers | 20–100 | 9 |
| Markup on model prices Token markup | None | Not published |
| Monthly cost per person Seat fee | None | None |
| GitHub stars GitHub stars | 6,109 | 409 |
| Quality testing Evals | Yes | Yes |
| Usage dashboards and logs Observability | Yes | Yes |
| Content guardrails Content guardrails | No | No |
| Prompt versioning Prompt management | Yes | Yes |
| Separate keys per team or app Virtual keys | Not published | Yes |
| SOC 2 audited SOC 2 audited | Yes | Yes |
| Will sign a HIPAA agreement HIPAA BAA | Yes | Yes |
| Can keep data in the EU EU data residency | Yes | Yes |
| You can export your request history Logs / usage data export | Yes | Yes |
Verified 3 days ago for Helicone and Verified 3 days ago for Braintrust Gateway. Want more fields, or a third option in the mix? Open these two in the full comparison tool.
Common questions
Is Braintrust Gateway free?
The Braintrust-hosted Gateway is free during public preview, with pricing to be announced before general availability. That means the eventual cost of running it is currently unknown, which is a planning risk for anything on your critical request path.
What is the difference between evals and observability here?
Braintrust is eval-centric — spans, scorers, datasets and CI quality gates that can block a deploy when output quality regresses. Helicone is production-analytics-centric — sessions, per-user breakdowns, custom properties and alerting on live traffic. They answer different questions and neither replaces the other.
Do either offer real-time guardrails?
No. Braintrust evaluates after the fact by design, and Helicone's guardrails and PII redaction are weaker than Portkey's or OpenRouter's. For inline blocking of prompts or responses, consider Portkey, Kong AI Gateway or Cloudflare AI Gateway.