Published Updated
In one or two sentences, before anything else.
AI/ML API (AIMLAPI) is a prepaid API service, contracting through Boiler Labs FZ-LLC in the UAE and also listing AIMLAPI OÜ in Estonia, that sells text, image, video and audio models from many makers through one key. Its current prices for OpenAI, Anthropic and Google models sit about 30% above the makers' list prices.
What it can do, one sentence each, and whether we actually tested it.
One prepaid balance for text, image, video, voice, music, embedding and 3D models, including GPT-6 Sol, Claude Opus 5.5, Gemini 4 Argon, Sora 2 and Veo 3.1.
AI/ML API claims 1000+ models; third-party listings have quoted 100+ to 400+ at different times.
An OpenAI-compatible API at api.aimlapi.com/v1, including /v1/responses, and an Anthropic-format endpoint that Claude Code can use through ANTHROPIC_BASE_URL.
Per AI/ML API's docs, only the Claude models it carries work through the Anthropic format.
mcp.aimlapi.com/mcp lets Claude Desktop, Claude Code, Cursor and other MCP clients search the catalogue, run text, image, video and audio models, manage long jobs and check the balance, with browser sign-in.
Per AI/ML API's MCP docs; calls are billed like the REST API.
Web search, page-to-Markdown, screenshots, news and social data callable with the same key over REST, function calling, MCP, or as an mcp tool inside a /v1/responses request for OpenAI models.
AI/ML API says only tool runs are billed; search and inspect calls are free.
The aimlapi CLI (npx aimlapi claude) signs you in, creates a key and starts Claude Code, Codex, OpenCode, Cline, OpenClaude, OpenClaw, Hermes Agent, zero or ZCode connected to AI/ML API.
Per its CLI docs; a positive balance is required.
Anthropic-style message batches for Claude models, returned within up to an hour.
AI/ML API says batching can cut costs by up to 50%; its docs list only Claude chat models for batches.
status.aimlapi.com (UptimeRobot) monitors six endpoints, including the API and the MCP server.
Endpoint pings, not per-model success rates.
The people it helps and the problems it solves, and where it ranks.
Built for
Problems it solves
Public tests other people ran: what they gave it and how it did.
Case 1
Input
export ANTHROPIC_BASE_URL=https://api.aimlapi.com and ANTHROPIC_API_KEY=<AI/ML API key>, set ANTHROPIC_MODEL to a Claude model, then run claude in the project; or run npx aimlapi claude.
What happened
Claude Code speaks the Anthropic format and AI/ML API accepts it, so nothing else changes. The guide notes that only Claude models work this way, and Claude Sonnet 5.5 costs 2.60 / 13 USD per million tokens here against Anthropic's 2 / 10 USD.
Case 2
Input
POST https://api.aimlapi.com/v1/responses with model gpt-4.1, the input "Find 3 fresh news items about the Model Context Protocol, with links.", and https://mcp.aimlapi.com/mcp as a tool of type mcp.
What happened
The model searches the tool catalogue, runs a news tool and answers in one request, and the output lists each mcp_call. This is the vendor's documented example; no third-party test of it has been published.
Case 3
Input
Text-to-speech requests to the VibeVoice model through AI/ML API's API.
What happened
The reviewer says requests timed out with HTTP 524 after about two minutes and a refund was refused; AI/ML API replied the next day. One user's report about one niche model, but refused refunds recur across its reviews.
What it does badly, with the evidence, and what we could not settle.
Weak
Prices above the makers' list prices.
Evidence: Its pricing page lists GPT-6 Sol at 2.60 / 13 USD per million tokens (OpenAI: 2 / 10), Claude Opus 5.5 at 5.20 / 26 USD (Anthropic: 4 / 20) and Claude Sonnet 4.6 at 4.1262 / 20.631 USD (Anthropic: 3 / 15), checked October 21, 2026.
No refunds and no free tier.
Evidence: The terms say the company "does not offer refunds for any products or services"; the docs FAQ says "The Free Tier is currently paused", so the first test costs a 20 USD top-up. Several Trustpilot reviewers report refused refunds.
Marketing the terms do not back.
Evidence: The home page advertises a "99.9% Uptime SLA" while the terms disclaim any guarantee of uptime or reliability and allow suspension without notice.
The privacy policy says nothing about prompts.
Evidence: It does not say whether prompts and outputs are logged, for how long, used for training, or which upstream providers receive them; no SOC 2 or ISO certification is named.
No control over routing.
Evidence: The docs describe no provider selection, fallback or load-balancing settings; AI/ML API chooses the upstream and the terms do not name it.
Not for buyers who watch cost or need contract terms.
Evidence: At about 30% over list with no refunds, no stated retention terms and an SLA its own terms disclaim, price-sensitive teams and anyone needing data or uptime commitments should use the makers' APIs or a list-price router such as OpenRouter.
Uncertain
How the two companies relate.
Why unsure: The terms name Boiler Labs FZ-LLC (UAE) as the contracting party, the footer lists AIMLAPI OÜ (Estonia), and disputes go to Estonian courts; no public document explains the link or which company takes payment.
Whether the model named is the model that answers.
Why unsure: A November 2025 Trustpilot reviewer says a model sold as GPT-5 answered as GPT-4; AI/ML API denies misrepresenting models, the reviewer's method was not published, and no third-party check exists.
Which upstream serves each model.
Why unsure: Neither the terms nor the docs say whether requests go to the makers' own APIs, cloud resellers or other sources.
How independent its online praise is.
Why unsure: Most Hacker News mentions are staff posts, and a staff member offered open-source projects a "50/50 revenue share" to add AI/ML API as a provider (GitHub, August and September 2026).
The numbers, each with its source and the date we checked it.
| Spec | Value | Note and source |
|---|---|---|
| Operator | Boiler Labs FZ-LLC, Ras Al Khaimah, UAE (contracting party); footer also lists AIMLAPI OÜ, Tallinn, Estonia | AIMLAPI OÜ (registry code 16931798) was registered on February 29, 2024; 2025 revenue 260,860 EUR, loss 22,765 EUR. Source: aimlapi.comchecked |
| Price vs makers' list prices | About 1.3 times list (GPT-6 Sol 2.60 / 13 USD per 1M tokens) | OpenAI lists GPT-6 Sol at 2 / 10 USD; Anthropic lists Claude Opus 5.5 at 4 / 20 USD against 5.20 / 26 USD here. Source: aimlapi.comchecked |
| Platform fee | None stated; the margin is in each model's price; 1 USD = 2,000,000 credits | Source: docs.aimlapi.comchecked |
| Minimum top up | 20 USD (Pay As You Go) | Source: aimlapi.comchecked |
| Payment methods | Card through Stripe (merchant of record); the home page plan card lists crypto | Crypto appears on the home page plan card; the terms name Stripe. Source: aimlapi.comchecked |
| Free tier | Paused: "The Free Tier is currently paused" | Source: docs.aimlapi.comchecked |
| Models | 1000+ claimed: text, image, video, voice, music, embeddings, 3D | Source: aimlapi.comchecked |
| API formats | OpenAI Chat Completions and Responses; Anthropic Messages for Claude models; Claude batches; function calling, streaming, web search on some models | The /v1/responses endpoint is shown in the Toolbox docs. Source: docs.aimlapi.comchecked |
| MCP | Official remote server at mcp.aimlapi.com/mcp (OAuth or API key): model search, inference, jobs, balance, Toolbox tools | Source: docs.aimlapi.comchecked |
| Agent Skills | No SKILL.md skill repository found; the docs publish an "Agent setup skill" page of instructions for coding agents | Source: docs.aimlapi.comchecked |
| Coding agents | aimlapi CLI for Claude Code, Codex, OpenCode, Cline, OpenClaude, OpenClaw, Hermes Agent, zero, ZCode; guides for Roo Code, Kilo Code, Aider; Cursor only via MCP | Source: docs.aimlapi.comchecked |
| Privacy and data retention | No statement on prompt logging, retention or training; no certifications named | Source: aimlapi.comchecked |
| Status page | API 100.000% over 30 days, 99.980% over 90 days; MCP 99.959% / 99.962% | Endpoint pings by AI/ML API's own monitor. Source: status.aimlapi.comchecked |
| Refunds | None: the company "does not offer refunds for any products or services" | Source: aimlapi.comchecked |
Answer first, then why, an example, the limits and what to do next.
AI/ML API costs more than the official APIs for current OpenAI, Anthropic and Google models, usually 1.3 times the makers' list prices. GPT-6 Sol is 2.60 / 13 USD per million tokens against OpenAI's 2 / 10 USD, and Claude Opus 5.5 is 5.20 / 26 USD against Anthropic's 4 / 20 USD.
Limits
Some older Claude models are 1.375 times list. Older directory copy claiming savings of up to 80% does not match the current price list.
AIMLAPI is a registered business with public terms and paying users, but its paperwork is split across two companies and its terms favour the seller. The terms name Boiler Labs FZ-LLC in a UAE free zone as the contracting party, the footer lists the Estonian AIMLAPI OÜ, and refunds are ruled out.
Why
AIMLAPI OÜ reported 260,860 EUR of revenue for 2025 to the Estonian register. Trustpilot shows 3.4 out of 5 from 23 reviews, split between five-star and one-star.
AI/ML API runs an official remote MCP server at https://mcp.aimlapi.com/mcp. MCP clients sign in through the browser or send an API key, then can search models, run text, image, video and audio generation, manage long jobs, check the balance and use its Toolbox of web tools.
Next step
In Claude Code: claude mcp add --transport http aimlapi https://mcp.aimlapi.com/mcp, then run /mcp and choose Authenticate.
AI/ML API's free tier is paused, so testing the API takes a top-up of at least 20 USD. The docs FAQ says "The Free Tier is currently paused", although the site still shows a Try For Free button and a few models carry a -free suffix.
Limits
Top-ups are not refundable under its terms.
Refunds are not offered under AIMLAPI's terms. The terms say the company "does not offer refunds for any products or services", and several Trustpilot reviewers report refused refunds, including after an accidental 20 USD charge.
OpenRouter costs less for the same OpenAI and Anthropic models, while AIMLAPI carries more video, image, music and 3D models under one key. OpenRouter bills at the provider's list price plus a 5.5% fee on credit purchases; AIMLAPI's prices run about 30% above list.
Limits
OpenRouter lets callers choose and order providers; AIMLAPI documents no routing controls.
This page is built from AI/ML API's own site, docs, terms, privacy policy and status page, the Estonian company register, the official price pages of OpenAI, Anthropic and Google, Trustpilot reviews, GitHub issues and Hacker News, all checked on October 21, 2026. AI/ML API's claims are labelled as its own.
AI/ML API (also written AIMLAPI) sells one prepaid key for other companies' models. Its pricing page claims "1000+ models" across text, image, video, voice, music, embeddings and 3D, including GPT-6 Sol, Claude Opus 5.5, Gemini 4 Argon, Grok 4.7, DeepSeek V4, Sora 2, Veo 3.1, Kling and Flux. Third-party directories have quoted 100+, 200+, 300+ and 400+ at different times. The balance is kept in credits at 1 USD = 2,000,000 credits, and Pay As You Go starts with a 20 USD top-up. The docs FAQ says "The Free Tier is currently paused", so there is no way to try the API before paying.
The API is OpenAI-compatible at https://api.aimlapi.com/v1. Claude Code works by pointing ANTHROPIC_BASE_URL at https://api.aimlapi.com, and a Batch API accepts Anthropic-style batches for Claude models. The docs describe no provider selection, fallback or load-balancing settings: AI/ML API picks the upstream.
For the current OpenAI, Anthropic and Google models checked, AI/ML API's prices are 1.3 times the makers' list prices. Some older Claude models are 1.375 times.
| Model, per million input / output tokens | AI/ML API | Maker's list price | Source |
|---|---|---|---|
| GPT-6 Sol | 2.60 / 13 USD | 2 / 10 USD | OpenAI |
| GPT-5.4 | 3.25 / 19.50 USD | 2.50 / 15 USD | OpenAI |
| Claude Opus 5.5 | 5.20 / 26 USD | 4 / 20 USD | Anthropic |
| Claude Sonnet 5.5 | 2.60 / 13 USD | 2 / 10 USD | Anthropic |
| Claude Sonnet 4.6 | 4.1262 / 20.631 USD | 3 / 15 USD | Anthropic |
| Gemini 4 Argon | 2.60 / 13 USD | 2 / 10 USD until November 30, 2026; 4 / 20 USD from December 1 |
AI/ML API prices come from its pricing page, which uses "$" without naming a currency. There is no separate platform or deposit fee; the margin is inside each price. Older directory copy repeats a vendor claim of saving "up to 80%" against OpenAI, which the current price list does not support. Video is priced per second (Sora 2 at 0.13 USD a second, Veo 3.1 at up to 0.78 USD a second); OpenAI's current pricing page no longer lists Sora, so that markup could not be compared.
No third-party latency, throughput or output-quality test of AI/ML API was found in public sources. What exists:
Trustpilot shows 3.4 out of 5 from 23 reviews (October 21, 2026), split between 52% five-star and 44% one-star. Praise covers the model range, same-day model additions, the playground and quick support. Complaints cover refused refunds (including an accidental 20 USD charge), a dashboard that showed over 135 USD for nine requests (the company called it a display bug), no way to test before paying, timeouts and "model overloaded" errors. Trustpilot notes the company replied to 75% of negative reviews, usually after more than a month.
Most Hacker News mentions are Show HN posts and comments by AI/ML API staff accounts. On GitHub, a staff member opened issues on open-source projects such as AutoGPT and Activepieces in August and September 2026, offering a "50/50 revenue share" for adding AI/ML API as a provider and claiming more than 400,000 users. Some provider integrations in open-source tools therefore come from paid outreach.
AI/ML API runs an official remote MCP server at https://mcp.aimlapi.com/mcp. Clients such as Claude Desktop, claude.ai, Cursor and Claude Code connect by browser sign-in (OAuth) or with an API key, and can then search and compare models, run text, image, video, audio and embedding models, manage long jobs and check the balance. The same server exposes its Toolbox: four tools (tools_search, tools_inspect, tools_run, tools_run_status) that reach web search, page-to-Markdown, screenshots, news and social data. Only tool runs are billed. Its Toolbox docs also show the MCP server passed as an mcp tool in a /v1/responses request for OpenAI models, so the model calls the tools inside one request.
For coding agents, the aimlapi CLI (npx aimlapi claude) signs you in, creates a key and starts Claude Code, Codex, OpenCode, Cline, OpenClaude, OpenClaw, Hermes Agent, zero or ZCode already pointed at AI/ML API. The docs also have guides for Roo Code, Kilo Code, Aider and continue.dev; Cursor appears only in the MCP guide. The Claude Code guide warns that only the Claude models AI/ML API carries work through the Anthropic format.
No SKILL.md skill repository was found. The docs publish an "Agent setup skill" page, which is a set of instructions for a coding agent to install the CLI. Official SDKs are aimlapi-sdk-python (built on the OpenAI SDK) and aimlapi-sdk-node, plus the langchain-aimlapi package; there is a LiteLLM guide and an llms.txt index of the docs.
The terms name Boiler Labs FZ-LLC, in a Ras Al Khaimah (UAE) free-zone business centre, as the contracting party. The site footer also lists AIMLAPI OÜ, an Estonian private company entered in the register on February 29, 2024 with a share capital of 1 EUR, and the terms put disputes under Estonian law and the Harju District Court. AIMLAPI OÜ reported revenue of 260,860 EUR for 2025 with a loss of 22,765 EUR, and 109,460 EUR for 2024. No public document explains how the two companies relate or which one receives payments.
| Service | Price against makers' list prices | Fee on top |
|---|---|---|
| AI/ML API | About 1.3 times | None stated |
| OpenRouter | Same | 5.5% on credit purchases |
| CometAPI | 0.8 times, by its own rule | None stated |
| ZenMux | Same | 5% on top-ups, waived for now |
| NanoGPT | Same | None on deposits |
OpenRouter also lets callers choose and order providers. AI/ML API's case rests on breadth (video, music, 3D and voice under one key), its MCP server and Toolbox, and the one-command CLI.
Anyone sensitive to price, anyone who may need a refund, and teams that need written data-retention terms, a contractual SLA or knowledge of which upstream serves each request. For trying many media models from one key with an MCP server attached, the main costs are the 30% premium and a non-refundable 20 USD top-up.
Sources
Changelog
APIMODELS is a pay-as-you-go AI API relay that sells about 150 models through one key at its own prices, mostly below the makers' list prices, and spends the margin on backup channels: 98.62% of requests succeeded over 30 days, moderation rejections excluded. APIMODELS is our own product: it is made by the team behind GenGrades.
ZenMux is an LLM gateway, operated under its terms by NexaMind Singapore Pte. Ltd., that sells Claude, GPT, Gemini and many Chinese models through OpenAI, Anthropic and Vertex-style endpoints at the makers' list token prices. It adds automatic credits for bad or slow answers, which it calls insurance, and 20 to 200 USD monthly coding plans.
WaveSpeedAI is an API platform, run by a Singapore company and a Hong Kong company named in its terms, that sells 1,000+ image, video, audio and 3D models plus an LLM service on one balance. It matches fal on Kling 3.0, charges above Google's list for Veo 3.1 Fast and accepts Alipay and WeChat Pay.
Vercel AI Gateway is Vercel's hosted API for about 480 text, image, video and speech models behind one key, billed from prepaid credits at the providers' list prices with no markup, including when you bring your own keys. It became generally available on August 21, 2025, and gives each Vercel team 5 USD of free credit a month.
Requesty is a London-based LLM gateway, run by Requesty Ltd and seed-funded by 20VC in 2025, that sells OpenAI- and Anthropic-compatible access to more than 600 models at list price plus a 5% top-up fee. It stands out for routing policies, spend controls and an EU endpoint in Frankfurt.
Replicate is an API for running AI models, owned by Cloudflare since a $57.4 million acquisition closed on December 1, 2025. It sells 100+ official models such as Veo 3.1, Nano Banana Pro and Kling 3.0 per output, and 50,000+ community models by the second of GPU time.