Together AI GPU Pricing
AI platform with dedicated GPU rental and GPU clusters alongside serverless inference.
| GPU | VRAM | On-demand /hr | Spot /hr | Reserved /hr | Per month | Notes |
|---|---|---|---|---|---|---|
| B300 · compare all providers | ||||||
| B300 (HGX) (min 8) | 288GB | $9.99 | $4.99 | — | $7,293 | GPU-cluster on-demand rate; spot = preemptible. Reserved: contact sales. |
| B200 · compare all providers | ||||||
| B200 | 180GB | $8.19 | $4.09 | $6.79 | $5,979 | GPU cluster on-demand; dedicated-inference endpoint $11.95/GPU-hr |
| H200 · compare all providers | ||||||
| H200 | 141GB | $5.99 | $2.99 | $3.99 | $4,373 | GPU cluster on-demand; dedicated-inference endpoint sales-gated |
| H100 · compare all providers | ||||||
| H100 | 80GB | $3.99 | $1.99 | $3.19 | $2,913 | GPU cluster on-demand; dedicated-inference endpoint $6.49/GPU-hr |
Per GPU-hour in USD, checked October 4, 2026. Monthly = 730 hours. Prices vary by region and availability; confirm on Together AI's pricing page.
Together AI GPU pricing FAQ
How much does B300 cost on Together AI?
Together AI lists B300 (B300 (HGX)) at $9.99 per GPU-hour on demand, $4.99 spot, about $7,293 per GPU per month running 24/7 (as of October 4, 2026).
How much does B200 cost on Together AI?
Together AI lists B200 (B200) at $8.19 per GPU-hour on demand, $4.09 spot, $6.79 reserved, about $5,979 per GPU per month running 24/7 (as of October 4, 2026).
How much does H200 cost on Together AI?
Together AI lists H200 (H200) at $5.99 per GPU-hour on demand, $2.99 spot, $3.99 reserved, about $4,373 per GPU per month running 24/7 (as of October 4, 2026).
How much does H100 cost on Together AI?
Together AI lists H100 (H100) at $3.99 per GPU-hour on demand, $1.99 spot, $3.19 reserved, about $2,913 per GPU per month running 24/7 (as of October 4, 2026).