<!--
Source: https://scrutica.com/facilities/fac-epoch-gc-together-ai-36k-gb200s
Format: native markdown twin rendered from the same data modules as the page (not an HTML conversion).
Machine access: https://scrutica.com/api/mcp (MCP) · https://scrutica.com/agents (all surfaces)
-->

# together.ai 36k GB200s

Facility record `fac-epoch-gc-together-ai-36k-gb200s` — data source: epoch-gpu-clusters (https://web.archive.org/web/20241120213534/https://www.together.ai/blog/nvidia-gb200-together-gpu-cluster-36k); authority tier 2; contains estimated values (is_estimated: true)

| Field | Value |
| --- | --- |
| Type | ai training |
| Status | announced |
| Location | US |
| Power capacity | 86.5 MW |
| IT load | not published |
| GPU count | 36,000 |
| Peak PFLOPS (FP16) | 90,000 |
| Owner | Together |
| Operator | not published |
| Announced | not published |
| Operational | not published |

## Cite

> Scrutica, "together.ai 36k GB200s" (facility record), https://scrutica.com/facilities/fac-epoch-gc-together-ai-36k-gb200s; data source: epoch-gpu-clusters; accessed 2026-07-27.

Full interactive record: https://scrutica.com/facilities/fac-epoch-gc-together-ai-36k-gb200s · Methodology for derived values: https://scrutica.com/methodology#flop-estimation · All programmatic surfaces: https://scrutica.com/agents