AI Video Generator
Create cinematic motion videos from text prompts or reference media
Prompt (Desired Motion & Scene)
English or Multilingual
0 / 1000
Advanced Settings (Negative Prompt & Random Seed)
AI Image Studio
Generate high-density visuals, transform compositions, or edit photos
Image Canvas
Your rendered AI image will appear here in high-resolution with zoom and download capabilities.
Generating Image with Agnes 2.5 Flash...
Chat Config
Agnes Reasoning LLM
Local Configuration & .env File System Manager
Built with HTML5 File System Access API — direct disk read/write with zero server required
Zero-Server Privacy Guarantee: Your Agnes API key and configuration settings are saved in your browser's private storage or written directly to your local file system as a
.env file. No database, server, or cloud proxy is involved.
Agnes AI Official Model Catalog & Documentation Reference
Current specs, endpoints, and billing rules grounded in Agnes AI documentation
| Modality | Model ID | Endpoint | Capabilities | Official Rate |
|---|---|---|---|---|
| Video | agnes-video-2.5-flash | POST /v1/videos | Text-to-video, Keyframe First/Last frame, Reference images | NOW $0 / sec |
| Video | agnes-video-2.5 | POST /v1/videos | 720P, 960P, 2K outputs, multimodal video & audio references | 720P: $0.025 /s |
| Image | agnes-image-2.5-flash | POST /v1/images/generations | 1K-4K resolution tiers, Base64/URL, Composition preservation | FREE $0 |
| Text | agnes-2.5-flash | POST /v1/chat/completions | General availability, reasoning extraction, code specialized | GA Tier |
| Text | agnes-2.5-pro | POST /v1/chat/completions | Scientific reasoning, long-context analysis, agentic workflows | Pro Tier |
Official Account Rate Limits (RPM) & Subscription Quotas
Grounded in official Agnes AI token plan specifications (tokenplan.md & code.md)
⚡ Rate Limit Rules & 429 Prevention:
- Shared Account Pool: Creating multiple API keys does not increase your total RPM. All keys under your account share the same limit pool.
- Video Generation Quota: Free/Default users have an effective limit of 1 video request per minute (1 RPM).
- Video Task Polling Cadence: Polling
GET /agnesapi?video_id=...faster than every 4-5 seconds triggers an HTTP 429 error. The dashboard includes built-in exponential backoff. - Reset Window: If you exceed your limit and hit an HTTP 429 error, wait 60 seconds for the rolling RPM window to reset.
🎬 Video Models RPM (All Video Endpoints)
| Account Tier | Peak RPM | Effective RPM | Video Quota |
|---|---|---|---|
| Free / Default | 2 RPM | 1 RPM (Max 1/min) | Standard |
| Enterprise Verified | 2 RPM | 2 RPM | High Priority |
| Token Plan User | 6 RPM | 5 RPM | 500 sec / day |
💬 Text & Reasoning Models RPM
| Account Tier | Allowed RPM | Effective RPM | Best Use Case |
|---|---|---|---|
| Free / Default | 30 RPM | 20 RPM | Basic testing & light chat |
| Enterprise Verified | 60 RPM | 40 RPM | Team development & coding |
| Token Plan User | 1,000 RPM | 1,000 RPM | Autonomous agent workflows |
🎨 Image Models RPM by Output Resolution Tier
| Resolution Tier | Sample Dimensions (1:1 / 16:9) | Free / Default Account | Enterprise Verified | Token Plan User |
|---|---|---|---|---|
| 1K Tier | 1024x1024 / 1312x736 | 20 RPM (30 peak) | 40 RPM (60 peak) | 100 RPM (120 peak) |
| 2K Tier | 2048x2048 / 2624x1472 | 10 RPM (20 peak) | 20 RPM (40 peak) | 80 RPM (120 peak) |
| 3K Tier | 3072x3072 / 3936x2208 | 1 RPM | 1 RPM | 1 RPM |
| 4K Tier | 4096x4096 / 5248x2944 | 1 RPM | 1 RPM | 1 RPM |
👑 Token Plan Subscription Quota Allowances
| Subscription Plan | Text Requests (agnes-2.5-flash) | Image Generation Quota | Video Generation Quota |
|---|---|---|---|
| Starter Plan | 1,500 req / 5 hrs (15,000 / week) | 4,000 images / day | 500 seconds / day |
| Plus Plan | 7,500 req / 5 hrs (75,000 / week) | 4,000 images / day | 500 seconds / day |
| Pro Plan | 30,000 req / 5 hrs (300,000 / week) | 4,000 images / day | 500 seconds / day |