Free AI Router · free LLM API radar | risk review of free AI API sites

// STATIONS 143 sites · page 4/6

Free public relayUp

vsllm.com

Source · routerpark

Overviewvsllm.com (Weiyun Model Open Platform) is a New API-based LLM relay-aggregation gateway, OpenAI-compatible. On sign-up it grants a daily auto-refreshing free quota (~20 calls / ~100K tokens per day) and advertises access to advanced models like Claude/GPT.

Free quotaAfter registration it offers a free subscription plan with quota that auto-refreshes daily without manual claiming (dashboard shows ~20 API calls and ~100K tokens/day, non-cumulative, reset next day); it also runs GitHub-star and daily-check-in reward activities.

Free modelsglm-5.3-flash-max gpt-5.6-terra glm-5.2-fast MiniMax-M2.7 auto claude-sonnet-5+34

See vsllm.com's free quota / risk review →vsllm.com

Commercial aggregatorUp

nexotoken.net

Source · routerpark

OverviewNexo (NexoToken, nexotoken.net) is a third-party commercial aggregation relay covering 130+ LLMs, with an OpenAI-compatible /v1 and a token console; its core is an ultra-low-multiplier (0.01x-0.03x) paid tier, with ~40 models offering a daily free quota (refreshing next day) as a lead-in.

Free quota~40 models offer daily free quotas that auto-refresh the next day (including some Claude/GPT/Gemini models); claims of sign-up bonus and 7-day no-questions refund are platform self-promotion — actual amounts per the console.

Free modelsClaude Sonnet 4.6/4.5 GPT-5.4 GPT-4o DeepSeek Kimi K2.6 GLM-5.1+3

See nexotoken.net's free quota / risk review →nexotoken.net

Commercial aggregatorUp

chutes.ai

Source · 12britz/awesome-free-models

OverviewChutes is a decentralized serverless AI inference marketplace on Bittensor (Subnet 64) with an OpenAI-compatible API hosting many open-source models; its early free tier is largely gone, now shifted to pay-per-token/subscription.

Free quotaThe early 'one-time $5 deposit for 200 requests/day' free tier has been removed; it is now pay-per-token with no subscription or minimum. Some models are near-free thanks to Bittensor TAO subsidies, but that depends on subsidies/miners and is unstable.

Free modelsDeepSeek-V3.2 Meta-Llama 3.1 Mistral

See chutes.ai's free quota / risk review →chutes.ai

Monitor/DirectoryUp

freellm.net

Source · 12britz/awesome-free-models

Overviewfreellm.net is a directory/navigation site for free LLM APIs. It catalogs 400+ free models across 30+ providers (Google AI Studio, Groq, NVIDIA NIM, OpenRouter and others) with their quotas, rate limits and signup links, and generates one-click config snippets for Claude Code, Cursor and Codex. It does not operate any callable inference endpoint of its own.

Free quotaThe site itself grants no free credits and issues no API keys. It only aggregates and displays the free tiers of third-party providers (e.g. Google AI Studio Gemini 2.5 Flash at roughly 500 requests/day, NVIDIA NIM at roughly 40 requests/minute, Groq with no credit card required). Users must register with each provider themselves to obtain keys.

Free modelsUnknown

See freellm.net's free quota / risk review →freellm.net

Monitor/DirectoryUp

hvoyai.com

Source · zzsting88/relayAPI

OverviewHvoy AI (hvoyai.com / hvoy.ai, mirrored domains) is a rankings, directory and diagnostics site for third-party AI API relay gateways. It uses real API keys to continuously measure uptime, latency, model-substitution ('adulteration') rate and pricing across gateways, but does not itself expose any programmatically callable free LLM API.

Free quotaThe site provides no free quota on an LLM API of its own. Its /free-tokens page distributes redemption codes and temporary keys donated by third-party relay gateways, gated behind a points/daily-lottery system (per community posts and search: roughly 20 points on signup, ~10 points per redemption, more points for sharing). Quota size, usable models and expiry are entirely set by the third-party gateway, and Hvoy explicitly disclaims any guarantee. A separate developer-facing 'pricing data API' returns gateway price data, not model inference.

Free modelsUnknown

See hvoyai.com's free quota / risk review →hvoyai.com

Monitor/DirectoryUp

hvoy.ai

Source · zzsting88/relayAPI

OverviewHvoy AI (hvoy.ai) is an AI API relay-station checker/price-comparison/ranking directory with a small 'free token' claim tacked on; it is not itself a stable, programmatically callable free LLM API service.

Free quotaThe site is primarily a checker/comparison tool; minor perks: ~20 points on signup, ~10 points to claim one free token, ~20 points for sharing, plus an occasional ~5 CNY redeem code; it claims a small daily 'freebie' of reverse-proxied sonnet-4.6/gpt5.3 (e.g. ~50 sonnet4.6 calls/day), with no guaranteed quota or stability.

Free modelssonnet-4.6（反代，白嫖有限）gpt5.3（反代，白嫖有限）

See hvoy.ai's free quota / risk review →hvoy.ai

Free public relayUp