// fair warning

Super simple.
Super flexible.

You do what the hell you want with it.

Careful.

The session, composed

Now, let's nerd out with session infrastructure.

At its simplest, cerver gives every user in your AI app a session — persistent, queryable, theirs. The model writing it and the compute running it are config; switch either one anytime, or run several in parallel on the same intent and curate the winner. One API.

$ curl -X POST https://gateway.cerver.ai/v2/sessions \
    -H "Authorization: Bearer $CERVER_API_TOKEN"
→ { "session_id": "sess_a4f7c9...", "status": "ready" }

Free at hobby scale. Free for the agents you build on it.

sess_a4f7c9… stays
refactor this auth flow
reading auth.py · 240 lines
✓ split into middleware.py + routes.py
now add tests
all of these can drive the session
model + CLI
Opus · Claude Code Sonnet · Claude Code Haiku · Claude Code GPT-5 · Codex CLI GPT-5 · OpenAI SDK Grok · xAI CLI Gemini · Google CLI Llama · open SDK
compute
E2B sandbox Vercel sandbox Cloudflare Workers local relay Modal Replicate Beam Cloud Daytona
— same id, same transcript, swappable underneath
The smallest claim

All your users' sessions. In one queryable place.

Every conversation a user has with your AI app is one row in this table. One user can own as many rows as your product gives them. The model writing each row and the compute running it are config you change per row. The table is the constant. One API.

sessions 1,247 rows
user session_id harness compute
user_4f3a sess_a4f7c9 Opus · CC E2B
user_8b21 sess_e2dc11 GPT-5 · Codex Vercel
user_4f3a sess_7791aa Sonnet · CC local
user_d019 sess_3b88e1 Grok · xAI Cloudflare
user_8b21 sess_b1227c Haiku · CC Modal new
+ 1,242 more · many per user

— same row shape, model + compute per user

Cerver

Cerver is the API for that layer.

Cerver gives you one interface for creating, running, switching, measuring, and closing AI sessions. But a good session layer should not only be reliable.

session apione interface
POST /v2/sessions
POST /v2/sessions/:id/run-llm
POST /v2/sessions/:id/compute
GET  /v2/sessions/:id/metrics
POST /v2/sessions/:id/close
Powerful sessions

Reliable is the baseline. Powerful is the point.

Cerver sessions are stable, transparent, and hard to accidentally break. They keep memory, tools, compute, billing, recovery, and user state together while your AI stack changes underneath. And they do more than hold state: they can switch models, move compute, track cost, recover runs, and spawn other sessions.

powerful session reliable core
switchmodels and compute
spawnparallel sessions
measurecost and latency
Create five versions with five intelligence layers.
Compare the outputs. Keep the best one. Same session record, clear cost trail.