Best for: Helicone is an Apache-2.0 open-source LLM observability platform that also offers gateway caching, rate limits and automatic fallbacks.
Categories › AI Gateway Software
Best AI Gateways for LLM Request Logging and Observability 2026
Helicone is the pick for logging every LLM request at the proxy, with an Apache-2.0 open-source core and a free Hobby plan of 10,000 requests. Portkey is the pick for a gateway and request logs in one plan at $49 a month, and LiteLLM is the pick for sending gateway logs to the tool you already run, such as Langfuse or OpenTelemetry. Cloudflare AI Gateway is the pick for free dashboard analytics with per-request logs on Cloudflare, and Langfuse is the pick for agent evals and observability next to any gateway.
Updated
Compare the deciding features
| Product | Documented buyer fit |
|---|---|
| Helicone | Helicone is an Apache-2.0 open-source LLM observability platform that also offers gateway caching, rate limits and automatic fallbacks.[1] [2] |
| Portkey | Portkey Developer plan is free forever with 10k recorded logs a month.[3] [4] [5] |
| LiteLLM | LiteLLM logging goes out to Langfuse, OpenTelemetry and GCS, S3 or Azure Blob buckets.[6] [7] [8] |
| Cloudflare AI Gateway | Cloudflare lists AI Gateway's core features as free, and they include dashboard analytics, caching and rate limiting.[9] [10] |
| Vercel AI Gateway | Vercel AI Gateway logs spend, model usage and observability metrics, including time to first token and input and output token counts.[11] [12] |
| Langfuse | Langfuse is an open source AI engineering platform for agent evals and observability, with a Hobby plan that is free and Core and Pro plans at $29 and $199 a month.[13] [14] [15] |
Ranked shortlist
Best for: Portkey Developer plan is free forever with 10k recorded logs a month.
Best for: LiteLLM logging goes out to Langfuse, OpenTelemetry and GCS, S3 or Azure Blob buckets.
Best for: Helicone, Portkey, LiteLLM and Cloudflare AI Gateway each record requests;
Best for: Vercel AI Gateway logs spend, model usage and observability metrics, including time to first token and input and output token counts.
Best for: LiteLLM logging goes out to Langfuse, OpenTelemetry and GCS, S3 or Azure Blob buckets.
Evidence and sources
- Helicone official page or repositoryRetrieved 2026-11-10T01:12:12Z
Open source LLM observability platform. One line of code to monitor, evaluate, and experiment.
- Helicone official page or repositoryRetrieved 2026-11-10T01:12:10Z
Gateway Caching Rate limits Automatic fallbacks
- Portkey official page or repositoryRetrieved 2026-11-10T01:12:10Z
Developer Free Forever
- Portkey official page or repositoryRetrieved 2026-11-10T01:13:01Z
Route to 250+ LLMs with 1 fast & friendly API
- Portkey official page or repositoryRetrieved 2026-11-10T01:13:00Z
MIT License Copyright (c) 2024 Portkey, Inc
- LiteLLM official page or repositoryRetrieved 2026-11-10T01:13:02Z
Log Proxy input, output, and exceptions using: Langfuse OpenTelemetry GCS, s3, Azure (Blob) Buckets
- LiteLLM official page or repositoryRetrieved 2026-11-10T01:12:10Z
One OpenAI-compatible API to 140+ providers and 1,800+ models.
- LiteLLM official page or repositoryRetrieved 2026-11-10T01:13:00Z
Virtual keys, budgets, RPM/TPM limits Spend by key, user, team & org Tag-based spend tracking Audit logs on every request
- Cloudflare AI Gateway official page or repositoryRetrieved 2026-11-10T01:12:10Z
AI Gateway's core features available today are offered for free, and all it takes is a Cloudflare account and one line of code to get started
- Cloudflare AI Gateway official page or repositoryRetrieved 2026-11-10T01:13:00Z
Each log can include the prompt, response, provider, timestamp, status, token usage, cost, duration, and user agent.
- Vercel AI Gateway official page or repositoryRetrieved 2026-11-10T01:13:02Z
The AI Gateway logs spend, model usage, and observability metrics related to your requests, which you can use to monitor and debug.
- Vercel AI Gateway official page or repositoryRetrieved 2026-11-10T01:12:12Z
AI Gateway charges no markup and no platform fee on tokens. You pay the provider's list price on a pay-as-you-go basis.
- Langfuse official page or repositoryRetrieved 2026-11-10T01:13:01Z
Langfuse is an **open source AI engineering platform** for **agent evals and observability**.
- Langfuse official page or repositoryRetrieved 2026-11-10T01:12:12Z
Hobby Get started, no credit card required. Great for hobby projects and POCs. Free
- Langfuse official page or repositoryRetrieved 2026-11-10T01:13:00Z
available under the "MIT Expat" license
1. Helicone: best for request logging captured at the proxy with an open-source core
Helicone is an Apache-2.0 open-source LLM observability platform that also offers gateway caching, rate limits and automatic fallbacks. Hobby is free with 10,000 requests and 1 GB of storage; Pro is $79 a month with unlimited seats and alerts; Team is $799 a month with SOC-2 and HIPAA compliance. Data retention runs 7 days on Hobby, 1 month on Pro, 3 months on Team and forever on Enterprise.
Sources: Helicone source · Helicone source
2. Portkey: best for logs and gateway in one plan at $49 a month
Portkey Developer plan is free forever with 10k recorded logs a month. The Production plan costs $49 a month and includes 100k recorded logs, with $9 per additional 100k requests, 30 days of log retention and 90 days of metrics. Portkey routes to 250+ LLMs through one API, and its gateway repository is MIT-licensed.
Sources: Portkey source · Portkey source · Portkey source
3. LiteLLM: best for sending gateway logs to the observability tool you already use
LiteLLM logging goes out to Langfuse, OpenTelemetry and GCS, S3 or Azure Blob buckets. LiteLLM is an open-source gateway that gives one OpenAI-compatible API to 140+ providers and 1,800+ models. It lists virtual keys, budgets, RPM/TPM limits, spend by key, user, team and organisation, tag-based spend tracking and audit logs.
Sources: LiteLLM source · LiteLLM source · LiteLLM source
4. Cloudflare AI Gateway: best for free dashboard analytics and request logs for teams on Cloudflare
Cloudflare lists AI Gateway's core features as free, and they include dashboard analytics, caching and rate limiting. Each log can include the prompt, response, provider, timestamp, status, token usage, cost, duration and user agent.
Sources: Cloudflare AI Gateway source · Cloudflare AI Gateway source
5. Vercel AI Gateway: best for spend and latency metrics for apps on Vercel
Vercel AI Gateway logs spend, model usage and observability metrics, including time to first token and input and output token counts. It charges no markup and no platform fee on tokens, and includes $5 a month of credit.
Sources: Vercel AI Gateway source · Vercel AI Gateway source
6. Langfuse: best for agent evals and observability next to any gateway
Langfuse is an open source AI engineering platform for agent evals and observability, with a Hobby plan that is free and Core and Pro plans at $29 and $199 a month. Hobby includes 50k units a month, 30 days of data access and 2 users, and the code is available under the MIT Expat licence.
Sources: Langfuse source · Langfuse source · Langfuse source
How it works
Gateway-level logging records each LLM request as it passes through the proxy: prompt, response, provider, token counts, cost and latency. This is different from general application performance monitoring, which traces services and infrastructure; see /software/observability/ for those tools. Langfuse is listed as a companion for agent evals and observability rather than as a gateway.
How to choose
Check what each log holds (prompt and response bodies, tokens, cost, latency) and how long it is retained on your plan. Check the allowance: Portkey counts recorded logs a month, Helicone counts requests, Langfuse counts units. Decide whether logs stay in the gateway's dashboard or export to your own stack (OpenTelemetry, S3, Langfuse).
Frequently asked questions
How do I log every prompt and response going to my LLM providers?
Put a gateway in the request path. Helicone, Portkey, LiteLLM and Cloudflare AI Gateway each record requests; a Cloudflare log can include the prompt, response, provider, token usage, cost and duration.
What does LLM request logging cost?
Helicone's Pro plan is $79 a month, Portkey's Production plan is $49 a month for 100k recorded logs, Langfuse Core is $29 a month, and Cloudflare AI Gateway's core features are free.
How long are LLM logs kept?
Portkey Production keeps logs 30 days and metrics 90 days. Helicone retains data 7 days on Hobby, 1 month on Pro and 3 months on Team.
Is an AI gateway the same as LLM observability?
A gateway such as LiteLLM routes requests and also logs them, while Langfuse is an open source platform for agent evals and observability.
Can gateway logs go to my existing observability stack?
LiteLLM logs to Langfuse, OpenTelemetry and GCS, S3 or Azure Blob buckets.
Methodology
Every price, limit, licence and feature comes from the vendor's or project's own pages, repository files and documentation, opened on 10 November 2026 (Europe/Helsinki), with the exact URL and a verbatim excerpt recorded for each fact. Prices are shown in the currency the vendor prints, with the billing period named. Picks follow documented fit for the buying situation and published facts, not paid placement or review-site scores. Vendor statements about compliance are quoted as the vendor states them. Vendor families are counted once.
How this comparison is made
Ranked on documented capabilities and verified pricing from official vendor sources. Every fact links to its source and check date, and the criteria are specific to this buying situation.
How we assessed this shortlist
Every price, limit, licence and feature comes from the vendor's or project's own pages, repository files and documentation, opened on 10 November 2026 (Europe/Helsinki), with the exact URL and a verbatim excerpt recorded for each fact. Prices are shown in the currency the vendor prints, with the billing period named. Picks follow documented fit for the buying situation and published facts, not paid placement or review-site scores. Vendor statements about compliance are quoted as the vendor states them. Vendor families are counted once.
Want your software in this comparison? A vendor profile listing costs $9.90 via Stripe and goes live within 24 hours of payment. Comparison-list placements cost a fee, and links are dofollow. Factual corrections are free.
Submit your software →Spot an error or a missing product?
Send a correction with a source link and we review it against the vendor's own documentation.
Suggest a correction