xAI Colossus API
Global Cloud Inference Host & API Provider · 3 Tracked Models
What is the latency, uptime SLA, and model lineup of xAI Colossus API?
xAI Colossus API maintains an average response latency of 210ms TTFT and an operational uptime SLA of 99.96% across its global infrastructure endpoints (us-east-1, us-central1). It currently hosts 3 tracked foundation models including Grok 3 Mini, Grok 4, Grok 3.
Verified daily via automated API latency tests and official documentation.
Zero-Shot RAG Verified Average TTFT Latency 210ms P90 Benchmark Window
Historical Uptime SLA 99.96% 30-Day Rolling Telemetry
Regional Deployment
us-east-1 us-central1
Multi-AZ Availability Models Hosted on xAI Colossus API (3)
Grok 3 Mini
xAI · Compact Reasoning ModelFast reasoning model delivering chain-of-thought code and math verification at low token cost.
Input / Out $0.30 / $1.20
Host TTFT 150ms
Throughput 160 t/s
Grok 4
xAI · Colossus Sparse MoE with Real-Time ReasoningxAI's flagship reasoning model trained on the 200k-GPU Colossus cluster, excelling at graduate-level science, code synthesis, and live web telemetry.
Input / Out $2.50 / $12.50
Host TTFT 480ms
Throughput 95 t/s
Grok 3
xAI · MoE TransformerLarge-scale xAI foundation model specialized in scientific reasoning, real-time data analysis, and enterprise logic.
Input / Out $3.00 / $15.00
Host TTFT 620ms
Throughput 88 t/s
Frequently Asked Questions About xAI Colossus API
What is the average latency and uptime SLA of xAI Colossus API?
xAI Colossus API delivers an average Time to First Token (TTFT) latency of 210ms and a verified 30-day rolling uptime SLA of 99.96%.
Which AI models are available on xAI Colossus API?
xAI Colossus API hosts 3 tracked LLMs on InferenceRate: Grok 3 Mini, Grok 4, Grok 3.
Which cloud regions does xAI Colossus API support?
xAI Colossus API supports API inference deployments across us-east-1, us-central1.