AXIOM
Get API Access

Model Specifications

The Model Family

Last updated

February 2026

Model Context Status Tier

AXIOM 2

200K tokens
Available
Latest

Our most capable and safest model

State-of-the-art performance on scientific reasoning, mathematics, and code — built with safety-first training from the ground up. Every capability advance is matched by rigorous safety evaluations.

Capabilities

Advanced scientific reasoning
Complex mathematical proofs
Production-grade code generation
Nuanced long-form analysis
52 language support
200K context window
Get access to AXIOM 2

AXIOM Research

1.0M tokens
Available
Research Preview

Frontier capabilities for research workloads

Designed for research teams requiring maximum capability on complex, multi-step reasoning tasks. Extended context, experimental features, and direct API access to our latest research models.

Capabilities

1M token context window
Extended reasoning mode
Research API access
Custom fine-tuning support
Priority compute allocation
Academic pricing available
Get access to AXIOM Research

AXIOM Instant

32K tokens
Available
Fast

Production speed for everyday tasks

Optimized for low-latency applications requiring fast, reliable responses. Built on our second-generation architecture with 5x the speed of AXIOM 2 at a fraction of the compute cost.

Capabilities

Sub-100ms response times
Competitive reasoning ability
High throughput API
Cost-optimized architecture
32K context window
Simple integration
Get access to AXIOM Instant

Developer API

Integrate in minutes.

Python, TypeScript, and REST. Full documentation, generous free tier, no rate limit surprises.

Get API key
example.py
import axiom

client = axiom.Axiom(api_key="your-api-key")

response = client.messages.create(
    model="axiom-2-20260101",
    max_tokens=4096,
    messages=[{
        "role": "user",
        "content": "Explain the implications of constitutional AI."
    }]
)
Get Started

Ready to build with frontier AI?

Access AXIOM 2 through our API. Production-grade safety, world-class capabilities, transparent pricing.