The AI Agent Framework That Ships in 8MB
Build, deploy, and scale AI agents in Go. No containers. No cold starts. No Python.
Get Started →From Prototype to Planet-Scale in 60 Seconds
1. Install
go get github.com/aixgo-dev/aixgo2. Create config/agents.yaml
supervisor:
name: coordinator
model: gpt-4o-mini # OpenAI - fast orchestration
max_rounds: 10
agents:
- name: data-producer
role: producer
interval: 1s
outputs:
- target: analyzer
- name: analyzer
role: react
model: claude-3-5-haiku # Anthropic - strong reasoning
prompt: |
You are a data analyst. Analyze incoming data and provide insights.
inputs:
- source: data-producer
outputs:
- target: logger
- name: logger
role: logger
inputs:
- source: analyzer3. Create main.go
package main
import (
"github.com/aixgo-dev/aixgo"
_ "github.com/aixgo-dev/aixgo/agents"
)
func main() {
if err := aixgo.Run("config/agents.yaml"); err != nil {
panic(err)
}
}4. Deploy anywhere
# Local development
go run main.go
# Production - single 8MB binary
go build -o agent
./agent
# Edge, Lambda, Cloud Run, Kubernetes - one binary, zero configurationWhy the Industry is Moving to Go
Python dominated AI because it was easy to prototype. Go will dominate production because it's built to ship.
Container Size
Python frameworks
1.2GB with dependencies
Aixgo
8MB single binary
Impact:
Deploy to edge devices, serverless, anywhere
Startup Performance
Python frameworks
30-45s cold start
Aixgo
<100ms instant startup
Impact:
True serverless viability, real-time response
Runtime Safety
Python frameworks
Runtime - Discover errors in production
Aixgo
Compile-time - Compiler catches errors before deploy
Impact:
Ship with confidence, sleep at night
LLM Data Validation
Python frameworks
Runtime only - Type changes found in production
Aixgo
Compile-time - Type changes caught before deploy, auto-retry
Impact:
Refactor with confidence, LLM errors auto-recover
The Go Advantage
| What Matters in Production | Python Frameworks | Aixgo |
|---|---|---|
| Deploy Anywhere | 1GB+ containers, complex deps | 8MB binary, zero deps |
| Cold Start Speed | 10-45 seconds | <100ms |
| Type Safety | Runtime discovery | Compile-time guarantees |
| Concurrency | GIL bottleneck | Native parallelism |
| Scaling Pattern | Rewrite for distribution | Same code, local → distributed |
| Operational Cost | High compute overhead | 60-70% infrastructure savings |
Built for What's Next
Aixgo isn't just another framework. It's the foundation for the next generation of AI systems.
13 Orchestration Patterns
All patterns production-ready: Supervisor, Sequential, Parallel, Router, Swarm, RAG, Reflection, Ensemble, and more. Go channels locally, gRPC for distributed.
8+ LLM Providers
OpenAI, Anthropic, Google Gemini, xAI, Vertex AI, HuggingFace, Ollama, and vLLM. Switch providers with a config change. Auto-detection by model name.
Full Observability Stack
OpenTelemetry tracing, Langfuse integration, Prometheus metrics, automatic cost tracking. Kubernetes health probes included.
Enterprise Security
4 auth modes, RBAC, SSRF protection, prompt injection defense, SIEM integration. Rate limiting and audit logging built-in.
Type-Safe Validation
Pydantic AI-style validation with automatic retry. 40-70% improvement in structured output reliability. Compile-time type checking.
60-70% Lower Infra Costs
8MB binaries, <100ms cold starts, ~50MB memory. Router pattern saves 25-50% on LLM costs. Local inference with Ollama for zero API costs.
What We're Building
- ✅ 8+ LLM providers (OpenAI, Anthropic, Gemini, xAI, Vertex AI, HuggingFace, Ollama)
- ✅ 13 orchestration patterns (Supervisor, Parallel, RAG, Reflection, MapReduce, etc.)
- ✅ Session persistence with JSONL and Redis backends
- ✅ Full observability (OpenTelemetry, Langfuse, Prometheus)
- ✅ Enterprise security (Auth, RBAC, TLS/mTLS, audit logging)
- ✅ Local to distributed with zero code changes
- 📋 Session encryption at rest
- 📋 PostgreSQL session backend
- 📋 Conversation branching and tree navigation
- 📋 Memory compaction and summarization
- 📋 Additional vector databases (Qdrant, pgvector)
- 📋 API stability guarantees and semantic versioning
- 📋 Kubernetes operator for automated lifecycle management
- 📋 Multi-modal support (Vision, Audio, Documents)
- 📋 Long-term memory and personalization
- 📋 Infrastructure as Code (Terraform modules)
Try v0.3.3
The AI infrastructure built on Python is reaching its limits. Aixgo is what comes next. Join us in building it.
Get Started →