这是一个关于免费的大模型api的合集，并精选了一部分模型
This is a collection of free LLM apis, and selected some models
我会尽可能更新维护这个项目（目前只有我一个人）
I will keep maintaining and updating this project to the best of my ability
入选原则是：限制请求速率而不是token > 尽可能多的来源 > 尽可能新且好的模型 > 足够用的请求速率
The selection criteria are: limit request rate over token count > more sources > newer and better models > sufficient rate limits
主要是有一定热度的文本模型
Primarily text models that have gained some popularity
欢迎大家分享更多api
Welcome to share more apis
这个表格是由Gemini 2.5 Pro生成的，由Taple渲染
This table was generated by Gemini 2.5 Pro, Rendered by Taple
| 名称 / Name | API | 模型 / Models | 请求速率 / Rate Limits | 后台 / Dashboard | 注 / Tips |
|---|---|---|---|---|---|
| ChatAnywhere | https://api.chatanywhere.tech |
gpt-4o-mini |
Not Limited | https://api.chatanywhere.org/ |
|
| 硅基流动 / SiliconFlow | https://api.siliconflow.cn/v1 |
deepseek-ai/DeepSeek-R1-0528-Qwen3-8B Qwen/Qwen3-8B THUDM/glm-4-9b-chat``THUDM/GLM-4-9B-0414``THUDM/GLM-Z1-9B-0414 THUDM/GLM-4.1V-9B-Thinking |
1000 RPM (each model) | https://cloud.siliconflow.cn/bills |
|
| OpenRouter | https://openrouter.ai/api/v1 |
deepseek/deepseek-r1:free deepseek/deepseek-r1-0528:free deepseek/deepseek-chat-v3-0324:free qwen/qwq-32b:free thudm/glm-z1-32b:free moonshotai/kimi-dev-72b:free openrouter/cypher-alpha:free tencent/hunyuan-a13b-instruct:free moonshotai/kimi-k2:free |
20 RPM / 200 RPD (each model) | https://openrouter.ai/activity |
|
| 书生 / Intern AI | https://chat.intern-ai.org.cn/api/v1 |
internlm3-latest |
10 RPM | https://internlm.intern-ai.org.cn/api/callDetail |
密钥有效期6个月 / The key is vailed for 6 months |
| 共享算力 / suanli.com | https://api.suanli.cn/v1 |
free:QwQ-32B |
Unknown | https://api.suanli.cn/detail |
算力由他人设备共享提供 / Shared computing by other people's devices |
| Google Gemini | https://generativelanguage.googleapis.com/v1beta/openai |
gemini-2.5-pro |
5 RPM / 100 RPD | https://console.cloud.google.com/apis/api/generativelanguage.googleapis.com/metrics |
|
| ↑ | ↑ | gemini-2.5-flash |
10 RPM / 250 RPD | ↑ | |
| ↑ | ↑ | gemini-2.5-flash-lite-preview-06-17 |
15 RPM / 1000 RPD | ↑ | |
| ↑ | ↑ | gemini-2.0-flash |
15 RPM / 200 RPD | ↑ | |
| MoeAPI | https://api.moeres.cn/v1 |
deepseek-ai/DeepSeek-R1-0528 deepseek-ai/DeepSeek-V3-0324 gemini-2.5-flash-lite-preview-06-17 gemini-2.5-flash-preview-04-17-thinking gemini-2.5-flash gemini-2.5-pro THUDM/GLM-4-32B-0414 THUDM/GLM-Z1-32B-0414 Qwen/QwQ-32B Qwen/Qwen3-30B-A3B Qwen/Qwen3-235B-A22B THUDM/GLM-4-32B-0414 gpt-4.1 gpt-4o claude-3-7-sonnet-20250219 claude-3-7-sonnet-20250219-thinking claude-opus-4-20250514 claude-opus-4-20250514-thinking doubao-seed-1-6-thinking-250615 doubao-seed-1-6-250615 ERNIE-4.5-300B-A47B grok-3-reasoner grok-4-nano grok-3 Kimi-Dev-72B o3-pro o4-mini tencent/Hunyuan-A13B-Instruct |
还没有开始限制 / Not Yet | https://api.moeres.cn/ |
稳定性差 / Poor stability |
| Cohere | https://api.cohere.ai/compatibility/v1 |
command-a-03-2025 |
20 RPM | https://dashboard.cohere.com/billing |
绑定支付方式可以使用速率限制跟宽松的Production Key / Binding payment methods can use rate limiting and relaxed Production Key |
| Bigmodel | https://open.bigmodel.cn/api/paas/v4/ |
GLM-4-Flash-250414 GLM-Z1-Flash |
只有并发数限制（均为30） / Only the number of concurrent transactions is limited (both 30). | ? | |
| Github Models | https://models.github.ai/inference |
openai/gpt-4.1-mini openai/gpt-4.1 |
15 RPM / 150 RPD | ? |
这是我的另一个项目，建议配套使用
My another project, recommended for use together
示例配置文件/Sample Configuration File : asak.json
asak: for-the-zero/asak
- llm_benchmark：个人评测榜单，可信度高，而且收录更全 / A personal review list, it is highly credible, and it is more comprehensive
- Artifical Analysis
- lmsys lmarena