Helicone is live · free plan, paid from $79/mo · last verified 9 Nov 2026 · viability healthy
Helicone
Helicone is an AI gateway and LLM observability platform that routes, logs, and caches traffic across 100+ model providers.
Buy Helicone when routing and the cost log should be one system, especially if a gateway taking a percentage of provider spend irritates you. The monitoring is competent and the pricing structure is honest. Skip it if deep eval and experiment tooling is your priority, and watch the ownership announcement before signing a multi-year Team deal.
Verdict generatedVerified 5d ago · liveness 87/100 · cite: rightaichoice.com/tools/helicone
- Teams routing production traffic across three or more LLM providers
- Companies that want routing, cost tracking, and alerts in one system
- Teams migrating off a gateway that takes a percentage of provider spend
- Engineering orgs enforcing rate limits and caching on every model call
- Teams needing a fully self-hosted stack — on-prem is Enterprise-only
- Hobby users exceeding 10,000 requests/month or needing retention beyond 7 days
- Projects prioritizing deep prompt evaluation and experiment tooling over gateway control
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip Helicone if your priority is systematic offline evaluation and experiment tracking rather than routing and cost visibility — its own comparison table marks that area limited against LangSmith and Braintrust.
Storage past the 1 GB free allowance is billed on usage — the calculator's 0.30 GB estimate runs $0.97/mo, so heavy log retention scales the bill
Hobby is genuinely free at 10,000 requests/mo, and Pro at $79/mo undercuts heavier observability suites once you account for 0% passthrough markup on provider costs. Team at $799/mo is roughly ten times Pro for five organizations plus SOC-2 and HIPAA — a steep step if you only need compliance, not org count. Compare against LangSmith and Braintrust, which go deeper on evaluation but are not gateways.
In short
Helicone — Helicone is an AI gateway and LLM observability platform that routes, logs, and caches traffic across 100+ model providers. Best for Teams routing production traffic across three or more LLM providers, Companies that want routing, cost tracking, and alerts in one system, Teams migrating off a gateway that takes a percentage of provider spend. Free to start; paid plans from $79/mo.
No head-to-head comparison for Helicone yet — the alternatives below are the closest matches we list.
What is Helicone?
Helicone — Helicone is an AI gateway and LLM observability platform that routes, logs, and caches traffic across 100+ model providers. Best for Teams routing production traffic across three or more LLM providers, Companies that want routing, cost tracking, and alerts in one system, Teams migrating off a gateway that takes a percentage of provider spend. Free to start; paid plans from $79/mo.
Helicone puts routing and monitoring in the same place. Point your code at its OpenAI-compatible endpoint and every request becomes a logged event you can query, cache, rate-limit, and fail over across providers. Passthrough billing keeps provider costs at 0% markup — the pricing calculator prices 10K requests at $0.00 in request fees, with storage driving the bill (about $0.97 for 0.30 GB). The daily surface is the observability side: request logs, segments, sessions, user analytics, custom properties, and HQL, a SQL-like language for slicing your own traffic. Alerts, reports, and webhooks push anomalies out; Playground and Datasets cover prompt iteration, and Prompt Management V2 lets you reference prompts by ID. The gateway adds caching, rate limits, and automatic fallbacks. Engineering teams running production traffic across several providers are the intended audience — the Pro tier adds unlimited seats, alerts, and HQL for $79/mo, Team $799/mo brings five organizations plus SOC-2 and HIPAA. The free Hobby tier is genuinely useful for a first project at 10,000 requests and 1 GB storage. Helicone's own comparison concedes that LangSmith and Braintrust go deeper on evaluation and experiments, while OpenRouter competes on routing alone. Note the ownership status: the site now states Helicone is no longer part of Mintlify, with details to come.
Key features
- AI gateway routing across 100+ LLM providers
- Passthrough billing at 0% markup on provider costs
- OpenAI-compatible API endpoint for any model
- Per-request LLM observability logging
- Sessions and user analytics for journey reconstruction
- Custom properties on request logs
- HQL SQL-like query language over request logs
- Playground with reasoning effort control
- Visual display of the model's thinking process where available
- Prompt Management V2 with version control
- Prompt-by-ID deployment without code rebuilds
- Datasets for versioning prompt data
- Caching to cut latency and provider spend
- Rate limits and automatic fallbacks on gateway traffic
- Alerts, reports, and webhooks on production LLM traffic
Behind the verdict
The pitch here is narrow and real: combine the gateway with the observability layer so you are not stitching Datadog, a proxy, and a cost spreadsheet together. We'd reach for this when a team already runs three or more providers and wants one place to see spend, cache hits, and failures without changing application logic much beyond the endpoint. Where it bites is scope. Helicone is a strong monitoring and routing tool that happens to have prompts and a playground. If your week revolves around experiment tracking, dataset scoring, and offline evals, LangSmith or Braintrust will feel more purpose-built. The site's own table says as much. Pricing is straightforward in a way most gateways are not. Passthrough billing means 0% markup on provider costs, and you pay for the platform, not a cut of your spend. The jump from Hobby (10,000 requests, 1 GB, 7-day retention) to Pro at $79/mo is where most teams land once retention and alerts matter. On compliance, Team brings SOC-2 Type II and HIPAA, and Enterprise adds SAML SSO and on-prem deployment. If a fully self-hosted stack is a hard requirement, that's an Enterprise conversation — not a mid-tier option. One honest caveat: the site's banner says Helicone is no longer part of Mintlify, with details to come. That's a roadmap and continuity question, not a product-quality one, but it's worth asking about before you commit to a long Team contract. Compared with OpenRouter, the routing overlap is real but Helicone's advantage is the logging and HQL underneath it. Compared with Langfuse, the differentiator is the gateway and passthrough billing more than the tracing UI. In practice we'd start a new project on Hobby, confirm the logs are useful, then move to Pro the month you cross 10,000 requests or need more than a week of
Models under the hood
as of 2026-11-08
Limitations
- The free Hobby plan caps at 10,000 requests per month, 10 logs/min ingestion, 60 API calls/min, 1 GB storage, and 7-day data retention.
- Pro ($79/mo) and Team ($799/mo) both carry usage-based pricing above the free allowance, so requests, storage, and ingestion can push the bill past the flat tier.
- Team raises ingestion to 15,000 logs/min and API access to 1,000 calls/min; Enterprise is the only tier with configurable retention, SAML SSO, 30,000 logs/min ingestion, and on-prem deployment.
- The site states Helicone is no longer part of Mintlify with details to come, so roadmap direction after the acquisition is unresolved.
as of 2026-10-21
Researching Helicone? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Is Helicone still active?
Helicone is live · free plan, paid from $79/mo · last verified 9 Nov 2026 · viability healthy.
Viability score
How well maintained and how widely used is Helicone? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: November 2026
How we score →What do users say about Helicone?
What people actually say about Helicone — is it worth it?
We ran a structured research pass across product reviews, community discussions, and post-purchase forum threads to surface the patterns vendors won't publish themselves. Below: the recurring strengths, the hidden costs people mention most, and the cohort that consistently regrets adopting this tool.
ResearchedDiscussed by users across 3 sources (Hacker News, GitHub, Lemmy) · researched Oct 16, 2026.
Weighted by the 34 posts each of 3 sources contributed.
- +Zero-markup passthrough billing means you pay provider rates, not a gateway tax
- +Per-request logging plus sessions and user analytics reconstruct full request journeys
- +Routing covers 100+ providers including Bedrock, Vertex, Groq, and OpenRouter
- +HQL lets power users run SQL-like queries over their own request logs
- +One-line integration is repeatedly cited as the fastest way to start monitoring
- −Now in maintenance mode after the Mintlify acquisition, so roadmap momentum is gone
- −Model and provider price data reportedly lags behind provider updates
- −No native per-feature cost rollups — trace-level data only, per community complaints
- −Open issue overstates cost for OpenAI Predicted Outputs due to a token math bug
- −170 open GitHub issues suggest a backlog larger than the team can clear
- • Passthrough billing carries 0% gateway markup, but re-routing through Helicone still adds a network hop you must operate and monitor
- • Self-hosting shifts hosting and maintenance cost onto your team, which is nontrivial for a ClickHouse-backed system
How we verified this
Every listed fact about Helicone — pricing, features, links, status — is re-read from the vendor's own pages by our verification pipeline and each change is logged with a date; the last pass ran on 9 Nov 2026. Analysis is drafted with AI models and checked against our editorial standards before publishing; see how we verify.
Verification history
We have re-verified Helicone 24 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — changed SSL certificate
- — changed pricing page
- — changed description, who should skip it, features, pricing tiers, who it suits, summary, our verdict
- — changed SSL certificate
- — changed link health
- — changed SSL certificate, description, pricing tiers, who should skip it, who it suits, our verdict
Showing the 6 most recent of 24 verification passes.
Free to cite with attribution — this page re-verifies continuously.
How much does Helicone cost in 2026?
Helicone: Free to start; paid plans from $79/mo. Pricing last verified 9 Nov 2026.
Plans compared
Every published Helicone tier, its price as the vendor publishes it, who it fits, and what it adds over the tier below. The 12-month cost projection follows.
| Plan | Price | Billing | Who it's for |
|---|---|---|---|
| Hobby | Free | — | A single engineer or side project making under 10,000 requests a month who wants logs and cost visibility without a credit cardAdds: Free entry point: 10,000 requests/mo, 1 GB storage, 1 seat, 7-day retention, community support |
| Pro | $79/mo | monthly | A growing engineering team that needs unlimited seats, alerts, and HQL to investigate production traffic togetherAdds: Adds unlimited seats, alerts and reports, HQL, 1-month retention, 1,000 logs/min ingestion, and 60 API calls/min over Hobby, with usage-based pricing on top |
| Team | $799/mo | monthly | A scaling company running several product lines that needs SOC-2 Type II and HIPAA coverage on its AI gatewayAdds: Adds 5 organizations, SOC-2 and HIPAA compliance, a private Slack channel, 3-month retention, and 15,000 logs/min ingestion over Pro |
| Enterprise | Custom | quote | A regulated or on-prem-constrained enterprise that needs SSO, custom contracts, and dedicated supportAdds: Adds custom MSA, SAML SSO, on-prem deployment, bulk cloud discounts, forever retention, 30,000 logs/min, and a dedicated support engineer with SLAs |
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Where the pricing makes sense
The company stage and team size where Helicone's pricing actually pencils out — and where peers do it cheaper.
Hobby is genuinely free at 10,000 requests/mo, and Pro at $79/mo undercuts heavier observability suites once you account for 0% passthrough markup on provider costs. Team at $799/mo is roughly ten times Pro for five organizations plus SOC-2 and HIPAA — a steep step if you only need compliance, not org count. Compare against LangSmith and Braintrust, which go deeper on evaluation but are not gateways.
Has Helicone's pricing changed?
No plan price for Helicone has moved since we began logging in Aug 2026; the pricing page is re-read on every verification pass.
Who should pick Helicone?
Pick Helicone if you are Teams routing production traffic across three or more LLM providers, Companies that want routing, cost tracking, and alerts in one system, Teams migrating off a gateway that takes a percentage of provider spend, Engineering orgs enforcing rate limits and caching on every model call.
Look elsewhere if you are Teams needing a fully self-hosted stack — on-prem is Enterprise-only, Hobby users exceeding 10,000 requests/month or needing retention beyond 7 days, Projects prioritizing deep prompt evaluation and experiment tooling over gateway control, Buyers who need a settled ownership and roadmap picture before committing long-term.
Skip it if Skip Helicone if your priority is systematic offline evaluation and experiment tracking rather than routing and cost visibility — its own comparison table marks that area limited against LangSmith and Braintrust.
GeneratedReal-world workflow fit
Concrete scenarios for the personas Helicone actually fits — and what changes day-one when you adopt it.
Point OpenAI-compatible requests at Helicone, turn on fallbacks so Anthropic picks up when OpenAI rate-limits, and watch per-request cost and latency in the dashboard
Outcome: One pane for routing and spend, with no percentage cut taken on provider costs and alerts firing when error rates spike
Move from Hobby to Pro for unlimited seats and HQL, write queries over request logs to attribute spend per customer using custom properties, then set reports for weekly cost review
Outcome: Cost attribution that used to be spreadsheet guesswork, and a shared URL every engineer can open when something breaks
Save a prompt with typed variables in the Playground, test reasoning effort at minimal versus high, version it, then reference the prompt ID through the gateway
Outcome: Prompt changes go live from the dashboard without a code deployment or rebuild
Use cases
- Monitor all LLM API calls in real time with detailed per-request logs and cost breakdowns
- Route requests across multiple providers to optimize for latency, cost, or reliability
- Debug failed requests and analyze user sessions to improve app performance
- Test prompts and models in the Playground with reasoning effort control before deploying
- Set alerts for rate limits, errors, or spending thresholds across your AI stack
- Track GPT-5, GPT-OSS, or Claude Opus 4.1 costs across Fireworks, Groq, and OpenRouter in one dashboard
- Ship prompt changes without code deployments using Prompt Management V2
- Filter request volume by country of origin to see where API calls come from
Setup time & first value
How long it actually takes to get something useful out of Helicone — broken out by persona, not the marketing-page minute.
One line of integration gets a request logged, so an individual engineer can see traffic in the dashboard in under 15 minutes. Pro's HQL queries and alerts take an afternoon to wire into a review routine. Team-level work (SSO, compliance review, five organizations) is an IT project measured in days, not minutes.
Switching to or from Helicone
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From OpenRouter: repoint your OpenAI-compatible base URL at Helicone's gateway and keep the same model strings, with 0% passthrough markup instead of a gateway cut
- →From direct provider SDKs: swap the base URL for Helicone's OpenAI-compatible endpoint — the changelog's one-line integration keeps your existing call structure
- →From LiteLLM: keep LiteLLM in the path as a documented integration and layer Helicone's logging and cost analytics on top
- ↗To LangSmith: export request data via the dashboard's data export option and rebuild evaluation and experiment workflows there
- ↗To Braintrust: move prompt iteration and scoring into Braintrust's experiment tooling, since Helicone's own comparison marks that area limited
- ↗To a direct provider setup: drop the base URL override and lose gateway caching, fallbacks, and unified cost reporting with it
Integrations
Resources & guides
Tutorials & Learning
YouTube returned 6 videos for “Helicone”, and we withheld 5: 5 could not be judged, because “Helicone” is a single word that other videos use for other things. Showing the 1 we can prove is about Helicone.
Official links
What are the best Helicone alternatives?
The closest alternatives to Helicone we list are Braintrust, LangSmith and LiteLLM. Ranked by direct product-type match, not category overlap.
| Alternative | Pricing | Starts at | Last verified |
|---|---|---|---|
| Braintrust | Freemium | Pricing not published | — |
| LangSmith | Freemium | Pricing not published | — |
| LiteLLM | Freemium | Pricing not published | — |
Alternatives to Helicone
View all
Braintrust
Braintrust delivers agent observability with evals, pattern discovery, and trace-scale storage in one platform.
Tools that pair well with Helicone
Common stack mates teams adopt alongside Helicone, with the specific reason each pairing earns its keep.
Braintrust
Braintrust delivers agent observability with evals, pattern discovery, and trace-scale storage in one platform.
LangSmith
LangSmith agent observability: trace, monitor, and evaluate every LLM run with SmithDB-backed trace search.
LiteLLM
LiteLLM is a self-hosted AI gateway that puts 140+ providers and 1,800+ models behind one OpenAI-compatible API.
Frequently asked questions
For makers of Helicone
- Get a Verified Listing for Helicone ($39): a written decision within 1 working day.
- Free listing for another tool: Free reviews are decided in order, within 3 weeks of the review starting.
Used Helicone? Help shape our editorial sentiment research.