Claude + Gemini
at your fingertips
One API key, two frontier model families — Claude(Opus 4.7, Sonnet 4.6, Haiku 4.5) & Gemini (3.1 Pro, 3 Flash, 3.1 Flash Lite). OpenAI-compatible. Free tier included.
Get Started
Login with Telegram to create your API key
Instant Setup
Login, grab your key, and start coding in under 30 seconds.
Claude + Gemini
Claude 4.x family via Anthropic API, Gemini 3.1 Pro / 3 Flash / 3.1 Flash Lite via OpenAI API.
Drop-in Compatible
Works with Claude Code, Cursor, Python & TypeScript SDKs — both formats.
5-Hour Rolling Limits
Limits reset automatically every 5 hours. No manual intervention required.
Zero Buffering Streaming
Full SSE streaming — Claude and Gemini. First tokens arrive instantly.
Per-Key Quotas
Split your pool across multiple keys. Usage logged per model.
Start in 3 steps
Replace your Anthropic base URL and you're live.
Get Your Key
Login via Telegram to receive your API key with budget and rate limits instantly configured.
Set Base URL
Point your client to https://opus.abhibots.com/v1 — everything else stays exactly the same.
Start Building
Use Claude Code, Python SDK, or cURL — it works seamlessly as a drop-in replacement.
curl -X POST https://opus.abhibots.com/v1/messages \
-H "x-api-key: ak-YOUR_KEY" \
-H "anthropic-version: 2023-06-01" \
-d '{"model":"claude-opus-4-7","max_tokens":100,
"messages":[{"role":"user","content":"Hello!"}]}'curl -X POST https://opus.abhibots.com/v1/chat/completions \
-H "Authorization: Bearer ak-YOUR_KEY" \
-d '{"model":"gemini-3.1-pro-preview",
"messages":[{"role":"user","content":"Hello!"}]}'Supported Models
All models tested and working ✓
Image Generation — now with Nano Banana 🍌
Two image models on the same key. Generate from the API POST /v1/images/generations or straight from the Telegram bot.
State-of-the-art image generation & editing — photoreal scenes, text rendering, and image-to-image. Use /nano in the bot.
/nano your promptFast, commercially-safe generation for clean product shots, logos and concepts. Use /image in the bot.
/image your prompt
Credit-based — buy a pack once, generate with either model. Firefly · 2 credits · Nano Banana · 5 credits.
Before You Buy
Try the free tier first — no commitment needed
Must Test Before Buying
This is a proxy service — latency will be higher than direct Anthropic API. Always test with the free tier first to see if it fits your workflow before purchasing a paid plan.
Works Great For
- ✓ Automation & bot workflows
- ✓ Normal AI chats & conversations
- ✓ Data analysis & research tasks
- ✓ OpenClaw / automated pipelines (no manual wait)
- ✓ Python scripts, background jobs
- ✓ Telegram & WhatsApp bots
Not Ideal For
- ✗ Heavy coding tasks (Claude Code, Cursor with large contexts)
- ✗ Latency-sensitive real-time apps
- ✗ Rapid back-and-forth IDE coding sessions
💡 Note: As a proxy, expect some added latency. It works but feels slower for interactive coding. For automated workflows & bots where you dont sit and wait — its perfect.
Plans
Simple pricing. Per 5-hour window. To buy, contact @flyingabhi on Telegram.
Enterprise
- 20M tokens / 5h
- Scale with +20M add-ons (₹7,999 each)
- 20 API keys · Dedicated support
Opus Proxy vs Official Anthropic
See how much you save — tokens reset every 5 hours
→ 530x cheaper than official pricing