🌐 English · 中文

aihubmix.com

Free LLM quota · free models · community reputation · risk review

Advertisement
TypeCommercial aggregator
StatusUp
Platformone-api
Reputation70 / 100
Risk level Med
Confidence75%
Discovered fromqixing-jk/all-api-hub

Overview

AIHubMix is a long-running commercial AI API aggregation relay (One API family) offering ~300 models (GPT/Claude/Gemini/DeepSeek/GLM, etc.) through an OpenAI-compatible endpoint on pay-as-you-go billing, plus a set of $0-priced free models subject to quotas.

Free quota

Sign up without a credit card; before the first top-up, all free models share 10 trial calls (never expire). After topping up ≥$1, free models are permanently usable at roughly 100 requests/day, 10 requests/min and 1M tokens/day (some heavier models capped at 33–50/day), reset daily; exceeding returns 429.

Advertisement

Free models

coding-glm-5.3-freecoding-kimi-k3-freecoding-minimax-m3-freexiaomi-mimo-v2.6-pro-freexiaomi-mimo-v2.6-flash-freeminimax-m2.7-freenemotron-3-ultra-550b-a55b-freenemotron-3-super-120b-a12b-freeqwen3.6-plus-preview-freeglm-4.7-flash-freeintern-s2-freehy3-freekimi-for-coding-free
🎁 Sponsored · free AI API credits & tools ↗

Pros

Cons

Risk flags

Community reputation

Several third-party relay review sites (eggstriker, apizhongzhuanzhan, zzznav, etc.) describe it as an established commercial relay with broad model coverage and easy integration, praising its permanent free testing quota and documentation; criticism focuses on mid-tier pricing and opaque stability data. A GitHub issue shows a user erroring because they had not switched to a free model name — a configuration issue. No exit-scam or major negative incidents were found; V2EX/Linux.do discussions show general concerns about relays' privacy and lack of consumer recourse (note: the 'AIHUB relay' on Linux.do is not confirmed to be the same entity).

Recommendation

Suitable for developers who want one key for many providers and some free large models for prototyping; the free tier requires a $1 top-up for long-term use and is tightly rate-limited. Avoid sending sensitive data; for production prefer official channels or keep multi-provider fallbacks.

Advertisement

Latest probe

ReachableYes (HTTP 200)
Latency733 ms
Platformone-api
Models endpoint300 (public)

Mentions (5)

aihubmix.com/models/free · neutral
Before first top-up: 10 trial calls shared across all free models; after $1+ top-up: 100/day, 10/min, 1M tokens/day, reset daily. ↗
eggstriker · neutral
Permanent free testing tier… mid-tier pricing… limited public data for independent verification of uptime claims; no formal SLA or DPA. ↗
apizhongzhuanzhan · positive
老牌一站式 AI API 中转,支持模型多、对接简单;大部分模型比官方稍贵,但提供大量免费测试额度。 ↗
GitHub · neutral
设置 AIHUBMIX_KEY 不会自动切换到免费模型,需显式配置 gpt-4o-free / glm-5 等免费模型名,否则报上游余额或权限错误。 ↗
V2EX · negative
对中转站整体的隐私与维权渠道表示担忧(泛中转讨论,非专指本站)。 ↗

Sources

Community discussion · reputation check

Want to verify aihubmix.com's real reputation before deciding? Read first-hand community discussion:

External discussion quality varies; judge together with the risk review on this page.

Advertisement

FAQ

Is aihubmix.com safe and trustworthy?

Overall reputation 70/100, risk level Med. Several third-party relay review sites (eggstriker, apizhongzhuanzhan, zzznav, etc.) describe it as an established commercial relay with broad model coverage and easy integration, praising its permanent free testing quota and documentation; criticism focuses on mid-tier pricing and opaque stability data. A GitHub issue shows a user erroring because they had not switched to a free model name — a configuration issue. No exit-scam or major negative incidents were found; V2EX/Linux.do discussions show general concerns about relays' privacy and lack of co…

Will aihubmix.com shut down? Is the service stable?

Known risks: Third-party relay: request content passes through the platform, posing data privacy and compliance risk; no public DPA; Free models are platform-subsidized; quotas and model list may change or be withdrawn at any time; Sustained free-tier use requires a $1 top-up, so it is not fully zero-barrier; Upstream key sourcing and stability are not independently verified; the claimed 99.99% availability lacks public data; Limited public information on the operating entity; unclear consumer recourse. Free and relay/public-benefit stations generally have no SLA and may shut down or change q…

Which free models does aihubmix.com offer? How much free quota?

Free quota: Sign up without a credit card; before the first top-up, all free models share 10 trial calls (never expire). After topping up ≥$1, free models are permanently usable at roughly 100 requests/day, 10 requests/min and 1M tokens/day (some heavier models capped at 33–50/day), reset daily; exceeding returns 429.. Available free models: coding-glm-5.3-free, coding-kimi-k3-free, coding-minimax-m3-free, xiaomi-mimo-v2.6-pro-free, xiaomi-mimo-v2.6-flash-free, minimax-m2.7-free, nemotron-3-ultra-550b-a55b-free, nemotron-3-super-120b-a12b-free, qwen3.6-plus-preview-free, glm-4.7-flash-free, i…

Is aihubmix.com worth using? How should I use it?

Suitable for developers who want one key for many providers and some free large models for prototyping; the free tier requires a $1 top-up for long-term use and is tightly rate-limited. Avoid sending sensitive data; for production prefer official channels or keep multi-provider fallbacks.

Related free-LLM sites

Advertisement
Advertisement
🎁 Sponsored · free AI API credits & tools ↗