AI, Made Simple.
Hamro AI is a free AI chat and gateway — one public key, dozens of free models across ten providers, routed with automatic failover. Ask questions, write code, or point your coding agent at it.
Here's a resilient fetch with exponential backoff:
import time, requests
def fetch(url, retries=3):
for attempt in range(retries):
try:
return requests.get(url, timeout=10).json()
except requests.RequestException:
if attempt == retries - 1:
raise
time.sleep(2 ** attempt) # backoffWant me to add rate-limit handling too?
54 free models, one key
Llama, Nemotron, DeepSeek, GLM and more — every request routes to a healthy provider automatically, so a hiccup never stops you.
Built for people and for coding agents
Native OpenAI and Anthropic protocols — the same key powers this chat, Claude Code, OpenCode, Cursor and any OpenAI SDK.
$ npm run claude # opens Claude Code on # Hamro AI's free models
export OPENAI_API_KEY=nishan-bajagain export OPENAI_BASE_URL="https://hamro.site/v1" opencode
curl -N https://hamro.site/v1/chat/completions \ -H "Authorization: Bearer nishan-bajagain" \ -H "Content-Type: application/json" \ -d '{"model":"random","messages": [{"role":"user","content":"hi"}], "stream":true}
Sticky-success routing picks the fastest healthy provider and remembers it.
429s, 5xx, timeouts and disconnects fall through the chain mid-stream.
Tool calling, code fences and SSE streaming pass through verbatim.
One public key, per-key namespaced history, no account required.
Start chatting — it's free
No signup, no credit card. Just open the chat and ask anything.
Open the chat