The easiest way to access LLM APIs for free β no credit card, no registration.
Copy a key from the table below, paste it into your app, and start building.
[](https://github.com/alistaitsacle/free-llm-api-keys/stargazers)
[](https://github.com/alistaitsacle/free-llm-api-keys/commits)
[]()
[]()
[](https://x.com/getkeyway)
**β Star this repo = more free keys for everyone**
**π [Follow @getkeyway on X](https://x.com/getkeyway) for fresh key drops, new model alerts & uptime status**
[English](./README.md) | [δΈζ](./README_CN.md) | [ζ₯ζ¬θͺ](./README_JA.md) | [νκ΅μ΄](./README_KO.md) | [EspaΓ±ol](./README_ES.md) | [PortuguΓͺs](./README_PT.md)
---
π§ͺ **[Try it in the Playground β](https://alistaitsacle.github.io/free-llm-api-keys/)** β paste a key, chat with 90+ models in your browser, no install
---
## π‘ Why This Project?
AI is changing everything, but most APIs are locked behind credit cards and paywalls. Not everyone has access β especially students, hobbyists, and developers in regions where these services aren't available.
I use LLM APIs heavily in my own work. I have spare capacity that would otherwise go to waste. So I decided to share it β free keys, no strings attached, updated daily.
**No credit card. No registration. Just copy, paste, and build.**
---
## πΌ Need higher availability?
For production-grade LLM keys, contact **alistaitscale@gmail.com**.
---
## π Available Keys
> β° Last updated: 2026-06-03 14:07 (UTC+8)
>
> All keys are API tokens compatible with **OpenAI SDK** β use them with any tool that supports custom OpenAI endpoints.
> New keys are published multiple times per day by the server cron. Expired keys are cleaned throughout the day. Each key has a budget ($20-$100) and expires in 24-48 hours.
>
> β οΈ **Note:** Keys are shared publicly. Budget may already be consumed by other users. If a key doesn't work, try another one or check back later for fresh keys.
>
> **[Verify your key here](https://alistaitsacle.github.io/free-llm-api-keys/)** β one-click check if a key still works.
### Featured models
GPT-5.5, Claude Opus 4.7, Gemini, DeepSeek, smart-chat, Kimi, image, audio, and embeddings.
The publisher only shows real keys. If a target model has no KM recommendation or quota, it tries a quota-backed KM-recommended model in the same capability; otherwise that shelf stays hidden.
### GPT-5.5 `06-03 14:07`
| Key | Model | Status | Budget | Rate Limit | Expires | Description |
|-----|-------|--------|--------|------------|---------|-------------|
| `sk-tZQfcbnjTVlzGtm7rHwaViBao1qHWt0GmY3oobijMy5Ia4pC` | gemini-2.5-pro | π New | $20 | 5 RPM | 2026-06-05 | KM recommended alternative for Premium GPT flagship |
| `sk-8NBcV3KyebZhw6TIkvDxNRUhtKK2FUrN9waGJ6MKs4H2mk3H` | gemini-2.5-pro | π New | $20 | 5 RPM | 2026-06-05 | KM recommended alternative for Premium GPT flagship |
| `sk-Fxfm1b7xaWqa48zgCaL7ajNmJGPKvKSDRygbDGeurTMzHmaY` | gemini-2.5-pro | π New | $20 | 5 RPM | 2026-06-05 | KM recommended alternative for Premium GPT flagship |
| `sk-RNasOfzhZbf0vhW0pKLjQ7BEShgq1CNvIPUPoyv2RNAd1w4W` | gemini-2.5-pro | π New | $20 | 5 RPM | 2026-06-05 | KM recommended alternative for Premium GPT flagship |
---
### Claude Opus 4.7 `06-03 14:07`
| Key | Model | Status | Budget | Rate Limit | Expires | Description |
|-----|-------|--------|--------|------------|---------|-------------|
| `sk-bgjLKLDImdTVfzGcd5kgWWHHKAJbVUPDMiCmZFfI3kmbiJlh` | claude-opus-4-7 | π New | $20 | 5 RPM | 2026-06-04 | Claude Opus flagship |
| `sk-3KC5WVYpY7gEKGLQ4yJfyB6TsbXs41KTfxmA7dgjmnDiKOKB` | claude-opus-4-7 | π New | $20 | 5 RPM | 2026-06-04 | Claude Opus flagship |
| `sk-pRmvcRsttaVM4w4LM9ifg7LDv1xRC7hYPDOSvMRe9N1fgBhi` | claude-opus-4-7 | π New | $20 | 5 RPM | 2026-06-04 | Claude Opus flagship |
| `sk-GiY1IsrIm4k2NHFFtMGBw7hA1aigFakPF6mkagMKAwhDgdfD` | claude-opus-4-7 | π New | $20 | 5 RPM | 2026-06-04 | Claude Opus flagship |
---
### Gemini `06-03 14:07`
| Key | Model | Status | Budget | Rate Limit | Expires | Description |
|-----|-------|--------|--------|------------|---------|-------------|
| `sk-nGxvWsz9RVf62OGbjiptiuCFMlBIbk5wShwErROP8JAS1VY8` | gemini-2.5-flash | π New | $20 | 20 RPM | 2026-06-04 | Fast Gemini option for long-context general chat |
| `sk-c7fEj0Pflh2WfuFJq9uUvQU9ETaanb8741CMfgUQkkc5LIaF` | gemini-2.5-flash | π New | $20 | 20 RPM | 2026-06-04 | Fast Gemini option for long-context general chat |
| `sk-nsuNAw5kUJ6kY6j0JvZ3BvGkiQBhbDrTjB3A952dU1Udomln` | gemini-2.5-flash | π New | $20 | 20 RPM | 2026-06-04 | Fast Gemini option for long-context general chat |
| `sk-UjGmzDTQiuhA1JQhDkeLwlRDuOyIkPxyxaMIoAolUr2pfZij` | gemini-2.5-flash | π New | $20 | 20 RPM | 2026-06-04 | Fast Gemini option for long-context general chat |
| `sk-UapGlJuQyvDboowuY890a3Qyqd5j5fMrKH2vAgt2v6a9u3Ct` | gemini-2.5-flash | π New | $20 | 20 RPM | 2026-06-04 | Fast Gemini option for long-context general chat |
| `sk-ncGnn0fCZ3BEmvow31TctGEhm7GttX7vN0dAksCsb2M1COh3` | gemini-2.5-flash | π New | $20 | 20 RPM | 2026-06-04 | Fast Gemini option for long-context general chat |
---
### DeepSeek `06-03 14:07`
| Key | Model | Status | Budget | Rate Limit | Expires | Description |
|-----|-------|--------|--------|------------|---------|-------------|
| `sk-PqZTvcsUCxbzMbdBtc24nAOpe3DXm4YLjST91sZFsbAJSa2S` | gemini-2.5-pro | π New | $20 | 20 RPM | 2026-06-04 | KM recommended alternative for Everyday chat, coding, translation, writing |
| `sk-ADscW2NDvYKkuwHvnsEw6IiDtcq7YzKdcOXmNsGwjmGBFyU2` | gemini-2.5-pro | π New | $20 | 20 RPM | 2026-06-04 | KM recommended alternative for Everyday chat, coding, translation, writing |
| `sk-ACB81N7ARULaC4fCTsxwM9j8U2IjNLKCdNe7VlRYLCHRTlK9` | gemini-2.5-pro | π New | $20 | 20 RPM | 2026-06-04 | KM recommended alternative for Everyday chat, coding, translation, writing |
---
### Kimi `06-03 14:07`
| Key | Model | Status | Budget | Rate Limit | Expires | Description |
|-----|-------|--------|--------|------------|---------|-------------|
| `sk-4fhUV3FpkGNSg6uJ72s3d4TMDi4iIZxLBoozpZmlRUressIY` | kimi-k2.5 | π New | $19 | 10 RPM | 2026-06-05 | Kimi long-context general model |
---
### Image / Audio / Embedding `06-03 14:07`
| Key | Model | Status | Budget | Rate Limit | Expires | Description |
|-----|-------|--------|--------|------------|---------|-------------|
| `sk-aG0NWMgYbRhXZRnb2YSZTb7oexpyB2FdEDPsRm4lj5SA1WzQ` | text-embedding-3-small | π New | $20 | 20 RPM | 2026-06-04 | Text embeddings |
| `sk-dfGKx3A1kPIMKu8Iyqslosv016IlArEtRoLQdvCh4Na8urMj` | text-embedding-3-small | π New | $20 | 20 RPM | 2026-06-04 | Text embeddings |
| `sk-Hl0QXYFh9wKj2X8sO3DVYkHqjNDA4AR4VsetJHXZHgBGuZUi` | text-embedding-3-small | π New | $20 | 20 RPM | 2026-06-04 | Text embeddings |
## π How to Use
All keys work with the **OpenAI API format**. Just set the base URL and paste your key.
**Base URL:** `https://aiapiv2.pekpik.com/v1`
### cURL
```bash
curl https://aiapiv2.pekpik.com/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer PASTE_KEY_HERE" \
-d '{
"model": "gpt-5.5",
"messages": [{"role": "user", "content": "Hello!"}]
}'
```
### Python
```python
from openai import OpenAI
client = OpenAI(
base_url="https://aiapiv2.pekpik.com/v1",
api_key="PASTE_KEY_HERE"
)
response = client.chat.completions.create(
model="gpt-5.5",
messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)
```
### Node.js
```javascript
import OpenAI from 'openai';
const client = new OpenAI({
baseURL: 'https://aiapiv2.pekpik.com/v1',
apiKey: 'PASTE_KEY_HERE',
});
const response = await client.chat.completions.create({
model: 'gpt-5.5',
messages: [{ role: 'user', content: 'Hello!' }],
});
console.log(response.choices[0].message.content);
```
### Image Generation (DALL-E 3)
```python
response = client.images.generate(
model="dall-e-3",
prompt="A cat astronaut floating in space, digital art",
size="1024x1024",
n=1
)
print(response.data[0].url)
```
### Text-to-Speech (TTS)
```python
response = client.audio.speech.create(
model="tts-1-hd",
voice="alloy",
input="Hello! This is a free text-to-speech demo."
)
response.stream_to_file("output.mp3")
```
### Embeddings
```python
response = client.embeddings.create(
model="text-embedding-3-small",
input="The quick brown fox jumps over the lazy dog"
)
print(f"Vector dimension: {len(response.data[0].embedding)}")
```
### Works With
| Tool | How to connect |
|------|----------------|
| **Cursor** | Settings β Models β API Key + Base URL |
| **ChatBox** | Settings β API β Custom Provider |
| **LobeChat** | Settings β Language Model β OpenAI |
| **NextChat** | Settings β OpenAI β API Key + Base URL |
| **Open WebUI** | Admin β Settings β Connections β OpenAI API |
| **Immersive Translate** | Settings β Translation β OpenAI |
| **Bob Translate** | Preferences β Services β Translation β OpenAI |
| **Any OpenAI-compatible app** | Set base URL to `https://aiapiv2.pekpik.com/v1` |
> More integration guides: [examples/guides/](./examples/guides/)
---
## π€ Supported Models
**90+ models** from 8+ providers, all accessible through a single API key.
| Provider | Models | Notes |
|----------|--------|-------|
| **OpenAI** | gpt-5.5, gpt-5.5-mini, gpt-5.5-nano, gpt-5.5-pro, gpt-4 | Latest GPT lineup |
| **Anthropic** | claude-opus-4-6, claude-sonnet-4-6, claude-haiku-4-5 | Latest Claude |
| **DeepSeek** | deepseek-chat (V3), deepseek-reasoner (R1) | Chat + reasoning |
| **Google** | gemini-3.1-flash-lite, gemini-3.1-flash-image | Gemini 3.1 |
| **xAI** | grok-4.20-beta, grok-4.20-multi-agent-beta | Latest Grok |
| **Mistral** | mistral-medium, codestral, devstral, mistral-small | Code + chat |
| **Cohere** | command-a-vision, command-a-reasoning, embed, rerank | Enterprise AI |
| **More** | Qwen, GLM-5, Mercury, MiniMax, Seed | Continuously expanding |
**Beyond chat:** Image generation (DALL-E 3), text-to-speech (TTS-1-HD), embeddings, reranking.
---
## β FAQ