MagnoAPI gives any AI app persistent memory across conversations. Two API calls. Works with Claude, GPT, Gemini — any LLM. No infrastructure to manage.
- Recommended by Surya S.B., Programmer and developer ★★★★★
Every request to your AI app follows this exact flow. MagnoAPI sits invisibly between your app and any LLM, giving it memory without changing how you call the LLM.
# ① Before calling your LLM — fetch relevant memories memories = requests.post( "https://magno-memory-api-production.up.railway.app/memory/search", headers={"X-API-Key": "magno_sk_your_key"}, json={"user_id": "user_123", "query": user_message, "top_k": 3} ).json()["results"] # Build enriched prompt context = "\n".join([m["text"] for m in memories]) prompt = f"Past context:\n{context}\n\nUser: {user_message}" # Call your LLM exactly as you normally would response = anthropic.messages.create(model="claude-sonnet-4-20250514", messages=[{"role":"user", "content": prompt}]) # ② After LLM responds — store this exchange requests.post( "https://magno-memory-api-production.up.railway.app/memory/store", headers={"X-API-Key": "magno_sk_your_key"}, json={"user_id": "user_123", "text": f"User asked: {user_message}"} )
// ① Before calling your LLM — fetch relevant memories const res = await fetch('https://magno-memory-api-production.up.railway.app/memory/search', { method: 'POST', headers: { 'Content-Type': 'application/json', 'X-API-Key': 'magno_sk_your_key' }, body: JSON.stringify({ user_id: 'user_123', query: userMessage, top_k: 3 }) }); const memories = (await res.json()).results; // Build enriched prompt const context = memories.map(m => m.text).join('\n'); const prompt = `Past context:\n${context}\n\nUser: ${userMessage}`; // ② After LLM responds — store this exchange await fetch('https://magno-memory-api-production.up.railway.app/memory/store', { method: 'POST', headers: { 'Content-Type': 'application/json', 'X-API-Key': 'magno_sk_your_key' }, body: JSON.stringify({ user_id: 'user_123', text: `User asked: ${userMessage}` }) });
# ① Search memories before your LLM call curl -X POST \ https://magno-memory-api-production.up.railway.app/memory/search \ -H 'X-API-Key: magno_sk_your_key' \ -H 'Content-Type: application/json' \ -d '{"user_id":"user_123","query":"your message","top_k":3}' # ② Store memory after your LLM responds curl -X POST \ https://magno-memory-api-production.up.railway.app/memory/store \ -H 'X-API-Key: magno_sk_your_key' \ -H 'Content-Type: application/json' \ -d '{"user_id":"user_123","text":"User asked about frameworks"}'
magno_sk_ key instantly. No credit card. No setup. Free forever up to 1,000 calls/month.POST /memory/search with the user's message. Get back the most relevant past memories. Inject them into your prompt.POST /memory/store after your LLM responds. The conversation is saved as a vector. Your AI gets smarter with every session.Free forever. No credit card. Ready in seconds.
By signing up you agree to the free tier limit of 1,000 calls/month. Upgrade anytime.