Serverless inference with pay-per-use token pricing ($0.0015 to $4.50 per 1M tokens for chat/vision models). Provisioned throughput with reserved PTUs for committed capacity. Dedicated GPU endpoints at $5.49-$8.99 per hour. Fine-tuning from $0.34-$40 per 1M tokens. GPU clusters at $1.99-$9.99 per GP
Not given
Key features
GPU clusters at scale with autoscaling
Voice agent infrastructure with sub-second latency
Comprehensive model library with 25+ regions
Offline verification in milliseconds
Cryptographic passports for agents
Revocation under 60 seconds
Authority classes and assurance tiers
No central authority calls required
Local registrar lifecycle
Request signing with short-lived keys
Public conformance testing
Fail-closed HTTP gating
Appeal mechanism for revocation
What makes it different
2x faster inference powered by cutting-edge research optimization
60% lower cost with workload-specific optimization
90% faster pre-training with Together Kernel Collection
Research-optimized infrastructure with proprietary kernels
Single unified API across multiple model providers
Neutral non-profit root barred by charter from issuing credentials or computing scores
Fully open standard with published conformance vectors and reproducible releases
Verification runs offline without phoning home to any authority
Four independent implementations agreeing on all conformance vectors
Public recorded ceremony to establish the root across independent custodians
Together AI
AINRA