A free, community-supported API service providing access to various AI models for developers and users.
初叶🍂Furry API offers 660 LLM API models.
API pricing per token ranges from $0.050 to $75.00/M (input).
Speed benchmark average: 171 tok/s.
https://ai.chuyel.topSecondary recommendation; prone to timeout errors (Claude often shows 524; use CloudFlare route for Claude)
https://ai.chuye.us.kgNot recommended; slower speed, but suitable as backup
https://ai.chuyel.cnOptimized route; faster access than others; recommended for priority use
Rankings are based on community-submitted tests and periodic health probes. Advisory only, not official data.
| Model | Speed | Latency | Tests |
|---|---|---|---|
264.87 tok/s | 8.57s | 5 | |
77.51 tok/s | 11.28s | 5 |
| Model | Audit | Input ($/M) | Output ($/M) |
|---|---|---|---|
baai/bge-m3 | — | $0.050 | $0.050 |
| — | $0.050 | $0.050 | |
qwen-turbo-2024-11-01 | — | $0.050 | $0.050 |
| — |
| Time | Model | Speed | Latency |
|---|---|---|---|
| May 21, 02:26 AM | gemini-3-flash | 264.87 tok/s | 8.57s |
| Apr 16, 12:34 PM | grok-4.20-beta | 77.51 tok/s | 11.28s |
This public welfare GPT site is provided by a certain individual. Free access and pricing are not under my control, please understand. If you lack Xiaoye coins, join the group to grab redemption codes or participate in a lottery.
As the public welfare site has reached 150 users, a lucky draw will be held to give away one E3 sub-account worth N yuan (5TB cloud storage + genuine Office365). To participate, please join the group. Draw time: April 27 at 15:00, participation is open until the draw.
Except for Gemini3f, all other gemini models have rate limits, and the rest are unknown.
Abuse and using third-party AI API sites are prohibited; otherwise, warnings or account bans may be imposed.
GPT can be used freely. Most Gemini models have rate limits, not imposed by this site. Grok currently has known rate limits: 39600 pure text chats or 19800 pure image generations per 20 hours.
| $0.050 |
| $0.050 |
nvidia/llama-nemotron-embed-1b-v2 | — | $0.050 | $0.050 |
sensenova-6.7-flash-lite | — | $0.050 | $0.050 |
qwen3-coder-30b-a3b-instruct | — | $0.050 | $0.050 |
nvidia/cosmos-reason2-8b | — | $0.050 | $0.050 |
| — | $0.050 | $0.050 |
THUDM/GLM-Z1-9B-0414 | — | $0.050 | $0.050 |
Server maintenance time: 2026/04/03 02:06, maintenance completed: 2026/04/02 04:34
Service has been restored, configuration upgraded from 4H8G to 4H16G. Compensation for downtime due to maintenance: one redemption code for a grab, and for timeout issues, two additional lucky draw redemption codes (all codes are exclusive to the group). Please understand. (All have been distributed)
The sign-in activity has ended. Thank you for your support.
The sign-in activity ended on April 1, and the maximum sign-in limit has been restored to 10 Xiaoye coins.
Suggestion to join the group: 644835603 (QQ group) to get the latest information.
huashang.dpdns.org
An OpenAI-compatible API relay providing access to multiple AI models with extensive model coverage and pricing data.
kfcv50.link
KFCV50 provides an OpenAI-compatible API relay for accessing AI models.
api2.aigcbest.top
Provides AI-generated content APIs for various applications, including text and image generation.
napi.seaya.link
Seamee API provides an AI model relay for accessing multiple LLMs through OpenAI-compatible endpoints.
new.waadri.top
WAADRI runs a unified AI model gateway that exposes aggregated model access through OpenAI-, Claude-, and Gemini-compatible interfaces.
api.newagiai.com
Provides API access to a wide range of AI models including text, image, and video generation from multiple providers.
