Comparison

Helicone alternative

Helicone is an open-source AI gateway and LLM observability platform. In March 2026, Mintlify acquired Helicone and said the product would continue in maintenance mode. Many teams now plan a move. This guide gives the current status, separates observability from routing, and compares the options.

Helicone status in November 2026

  • 3 March 2026: Mintlify announced that it acquired Helicone. The Helicone team joined Mintlify.
  • Maintenance mode: Mintlify wrote that Helicone will continue in maintenance mode, with security updates, bug fixes and new models.
  • Migration help: Mintlify wrote that it will work with every customer on a move to another platform.
  • New notice: the Helicone website now says that Helicone is part of Experiential Labs, with details to come. We could not find more official details on 20 November 2026.
  • Plans still listed: the pricing page shows a free Hobby plan with 10,000 requests per month and 7-day retention. It also shows paid Pro and Team plans.
What this means: the hosted service still runs, but no new features are planned. The code is open source under Apache 2.0, so you can also self-host it. For a long-term setup, plan a move while the service still works.

Observability or routing: what do you need to replace?

Helicone does two jobs. Find out which job you use before you choose an alternative.

JobWhat it coversTypical alternatives
ObservabilityRequest logs, prompt and response views, cost per request, sessions, evaluationsLangfuse, Portkey, your own logging
Routing (AI gateway)One OpenAI-compatible endpoint, fallbacks between providers, one keyLiteLLM, OpenRouter, Portkey, FreeLLMAPI
BothA gateway that also logs each requestPortkey, or a gateway plus Langfuse

Many teams split the two jobs. A gateway sends the requests, and a separate tool stores traces and evaluations. LiteLLM, for example, has callbacks that send logs to Langfuse, MLflow and other tools.

Helicone alternatives compared

Helicone alternatives (checked 20 November 2026)
ToolMain jobWhere it runsStatus and license
LangfuseObservability and evaluations for LLM apps and agentsHosted, or self-hosted with Docker Compose, Helm or TerraformMIT, except the ee folders. Part of ClickHouse since January 2026
LiteLLMAI gateway with virtual keys, spend tracking and fallbacksSelf-hostedOpen source. Logs to other tools with callbacks
PortkeyHosted AI gateway with logsHostedFree Developer tier, not for production. Paid Production tier
OpenRouterHosted model gateway with one billHostedFree models with daily caps, paid credits for the rest
FreeLLMAPISelf-hosted gateway that pools free provider tiersYour computer or serverOpen source (MIT). Single-user
Helicone (self-hosted)Gateway and observabilityYour server with Docker ComposeApache 2.0. Maintenance mode upstream

For details on the gateways, read LiteLLM alternative, Portkey alternative and free LLM API gateways compared.

What FreeLLMAPI replaces and what it does not

FreeLLMAPI replaces the gateway job, with a different aim. It routes each request only to provider free tiers, so inference costs $0. It records basic data for each request. It is not a full observability platform.

  • Per request: provider, model, status, input and output tokens, latency and time to first token.
  • Per failed attempt: the provider, the error class and a short redacted error message.
  • Dashboard: p50 and p95 latency, time to first token, success rate and token use over 24 hours to 90 days.
  • Headers: X-Routed-Via shows the provider that answered. X-Fallback-Trail shows each provider that failed.
  • Not included: prompt management, evaluations, user sessions or a team workspace. Use Langfuse or a similar tool for these.
Strongest free models with tool calling, today
ModelFree onContextFree limits (best offer)ToolsVision
GLM 5.3UnoRouter1M1 req/min
1 req/min per model
YesNo
Qwen3.8 MaxACLIDE-€20/month shared internal usage credits · no cardYesNo
GLM 5.3 FlashNVIDIA NIM, UnoRouter1M40 req/min
40 RPM
YesYes
Gemini 3.8 FlashGoogle AI Studio1MRecurring provider quotaYesYes
Gemini Flash (latest alias)Google AI Studio-10 req/min
20 req/day
250,000 tokens/min
~3M
YesYes
Qwen3.8 Flash NextAMD Radeon Cloud262KRecurring daily account allowanceYesNo
Qwen3.8 FlashElectronHub1M$0.25/week shared recurring creditYesYes
Gemini 3.7 FlashGoogle AI Studio1M10 req/min
20 req/day
250,000 tokens/min
~3M
YesYes

Showing 8 of 146 matching free models. See all models.

Live data from the FreeLLMAPI catalog (version 2026.11.19, checked 19 November 2026). Free limits change often. The catalog is checked every day.

Checklist before you choose

  1. List the Helicone features that your team uses each week. Most teams use only a few.
  2. Decide if you need a hosted service or a self-hosted tool.
  3. Check where each tool stores your prompts and responses. Some teams must keep data in one region.
  4. Check the retention period on the free plan of each tool.
  5. Check if the tool can send data to a second tool, for example a gateway that logs to Langfuse.
  6. Run the new setup next to Helicone for a short time before you switch off Helicone.

If your main cost is inference, not logging, look at the gateway first. A gateway that uses free provider tiers can lower the cost of tests and development to $0. Keep paid providers for the traffic that needs them.

How to move off the Helicone gateway

The Helicone gateway is OpenAI-compatible, with the base URL https://ai-gateway.helicone.ai. Most apps change two values to move. Do these steps:

  1. Export the Helicone data that you want to keep, before you change anything.
  2. Choose the new gateway and the new observability tool.
  3. Change the base URL and the API key in your app.
  4. Change the model names to the names of the new gateway.
  5. Send test traffic and compare the logs with Helicone for one day.
  6. Remove the Helicone key from your app.

Before and after: change the base URL and the key

from openai import OpenAI

# Before: Helicone AI Gateway
# client = OpenAI(base_url="https://ai-gateway.helicone.ai", api_key="<helicone-key>")

# After: FreeLLMAPI on your own computer
client = OpenAI(base_url="http://localhost:3001/v1", api_key="<unified-key>")

resp = client.chat.completions.create(
    model="auto",
    messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content)
Use every one of these through one key. FreeLLMAPI is an open-source, self-hosted router that puts all these free tiers behind a single OpenAI-compatible endpoint and fails over when one is rate-limited. Browse the catalog or go live.

Frequently asked questions

Is Helicone shutting down?

No shutdown date was announced. Mintlify acquired Helicone in March 2026 and said Helicone will continue in maintenance mode, with security updates, bug fixes and new models.

Who owns Helicone now?

Mintlify announced the acquisition on 3 March 2026. The Helicone website now also says that Helicone is part of Experiential Labs, with details to come. We could not confirm more details.

What is the best open-source Helicone alternative?

For observability, Langfuse is open source and self-hostable. For routing, LiteLLM is open source and self-hosted. FreeLLMAPI is an open-source gateway that routes only to free provider tiers.

Can I self-host Helicone?

Yes. The Helicone repository is Apache 2.0 licensed and includes a Docker Compose file. Upstream development is in maintenance mode, so plan for fewer updates.

Does FreeLLMAPI replace Helicone observability?

Partly. It logs status, tokens, latency and failed attempts for each request, with a dashboard. It has no prompt management, evaluations or team workspace.

Sources

All guides · Browse all free models

Keep reading

All posts →
Comparison

FreeLLMAPI vs OpenRouter

How the two compare for free access, limits, and key privacy.

Comparison

An OpenRouter free alternative

Hitting OpenRouter's 50/day free cap? The self-hosted alternative with no cap.

Comparison

FreeLLMAPI vs Portkey

Free vs a $49/month Production tier — pricing, key storage and log retention.