Together AI
Together AI sells open models per token through a serverless API and rents NVIDIA GPU clusters by the hour, on demand or reserved.
- Website: https://www.together.ai
- Headquarters: San Francisco, United States
- Kind: GPU cloud, Per-token API
Together AI sells 7 of the compared models per token and prices 3 kinds of GPU for hosting your own.
Models sold per token
| Model | Input, per 1M tokens | Output, per 1M tokens |
|---|---|---|
| DeepSeek V4 Pro 0813 | $1.32 | $3.96 |
| DeepSeek V4.1 Flash | $0.30 | $1.20 |
| GLM 5.2 | $1.40 | $4.40 |
| GLM 5.3 | $1.40 | $4.40 |
| GLM 5.3 Flash | $0.15 | $0.50 |
| Kimi K3 | $3.00 | $15.00 |
| OpenAI gpt-oss 120B | $0.15 | $0.60 |
GPUs for hosting your own
| GPU | From, per GPU-hour | Cheapest instance | GPUs in it |
|---|---|---|---|
| H100 | $3.99 | hgx-h100.8gpu | 8 × 80 GB |
| H200 | $5.99 | hgx-h200.8gpu | 8 × 141 GB |
| B200 | $8.19 | hgx-b200.8gpu | 8 × 180 GB |
Where it runs
One global price; Together AI chooses where a request is processed.
No place stated for 7 models: one global price, and Together AI does not say where a request is processed.
Where the prices come from
- Per-token prices: docs.together.ai/docs/serverless/models
- GPU servers: www.together.ai/pricing
Sources
List prices collected 4 November 2026, refreshed every three hours. No negotiated rates or taxes.