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.
Observability or routing: what do you need to replace?
Helicone does two jobs. Find out which job you use before you choose an alternative.
| Job | What it covers | Typical alternatives |
|---|---|---|
| Observability | Request logs, prompt and response views, cost per request, sessions, evaluations | Langfuse, Portkey, your own logging |
| Routing (AI gateway) | One OpenAI-compatible endpoint, fallbacks between providers, one key | LiteLLM, OpenRouter, Portkey, FreeLLMAPI |
| Both | A gateway that also logs each request | Portkey, 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
| Tool | Main job | Where it runs | Status and license |
|---|---|---|---|
| Langfuse | Observability and evaluations for LLM apps and agents | Hosted, or self-hosted with Docker Compose, Helm or Terraform | MIT, except the ee folders. Part of ClickHouse since January 2026 |
| LiteLLM | AI gateway with virtual keys, spend tracking and fallbacks | Self-hosted | Open source. Logs to other tools with callbacks |
| Portkey | Hosted AI gateway with logs | Hosted | Free Developer tier, not for production. Paid Production tier |
| OpenRouter | Hosted model gateway with one bill | Hosted | Free models with daily caps, paid credits for the rest |
| FreeLLMAPI | Self-hosted gateway that pools free provider tiers | Your computer or server | Open source (MIT). Single-user |
| Helicone (self-hosted) | Gateway and observability | Your server with Docker Compose | Apache 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-Viashows the provider that answered.X-Fallback-Trailshows each provider that failed. - Not included: prompt management, evaluations, user sessions or a team workspace. Use Langfuse or a similar tool for these.
| Model | Free on | Context | Free limits (best offer) | Tools | Vision |
|---|---|---|---|---|---|
| GLM 5.3 | UnoRouter | 1M | 1 req/min 1 req/min per model | Yes | No |
| Qwen3.8 Max | ACLIDE | - | €20/month shared internal usage credits · no card | Yes | No |
| GLM 5.3 Flash | NVIDIA NIM, UnoRouter | 1M | 40 req/min 40 RPM | Yes | Yes |
| Gemini 3.8 Flash | Google AI Studio | 1M | Recurring provider quota | Yes | Yes |
| Gemini Flash (latest alias) | Google AI Studio | - | 10 req/min 20 req/day 250,000 tokens/min ~3M | Yes | Yes |
| Qwen3.8 Flash Next | AMD Radeon Cloud | 262K | Recurring daily account allowance | Yes | No |
| Qwen3.8 Flash | ElectronHub | 1M | $0.25/week shared recurring credit | Yes | Yes |
| Gemini 3.7 Flash | Google AI Studio | 1M | 10 req/min 20 req/day 250,000 tokens/min ~3M | Yes | Yes |
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
- List the Helicone features that your team uses each week. Most teams use only a few.
- Decide if you need a hosted service or a self-hosted tool.
- Check where each tool stores your prompts and responses. Some teams must keep data in one region.
- Check the retention period on the free plan of each tool.
- Check if the tool can send data to a second tool, for example a gateway that logs to Langfuse.
- 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:
- Export the Helicone data that you want to keep, before you change anything.
- Choose the new gateway and the new observability tool.
- Change the base URL and the API key in your app.
- Change the model names to the names of the new gateway.
- Send test traffic and compare the logs with Helicone for one day.
- 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)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.
Related guides
- LiteLLM alternative
- portkey alternative
- free llm api gateways compared
- freellmapi vs litellm
- Free LLM API for AI agents
Sources
- Mintlify blog: Mintlify acquires Helicone (checked 20 November 2026)
- Helicone website (Experiential Labs notice) (checked 20 November 2026)
- Helicone pricing (checked 20 November 2026)
- Helicone docs (checked 20 November 2026)
- Helicone GitHub repository (checked 20 November 2026)
- Langfuse GitHub repository (checked 20 November 2026)
- FreeLLMAPI docs: Observability (checked 20 November 2026)