TensorX API pricing
11 TensorX models. Source: https://tensorx.ai/pricing
- Provider type
- Inference host
- Headquarters
- Ireland (EU)
- Regions
- EU
- EU data residency
- yes
- Residency notes
- Pricing page: all inference on EU-sovereign infrastructure with zero data retention; 'EU data residency out of the box'. Dedicated clusters in Dublin, Helsinki and Paris regions. TensorX Ltd., Dublin, Ireland (formerly Tensorix).
- Residency source
- https://tensorx.ai/pricing
HQ country is from public company information; regions and EU residency are only filled from the cited source page, otherwise 'unknown'.
Blend input:output
| Provider | Model | Input | Output | Cached in | Blended 3:1* ▲ | Context | EU residency |
|---|---|---|---|---|---|---|---|
| TensorX Inference host | DeepSeek V4 Flash 0731 | $0.25 | $0.30 | $0.06 | $0.263 | 1M | yes |
| TensorX Inference host | GLM-5.3 Flash | $0.20 | $0.50 | $0.05 | $0.275 | 1M | yes |
| TensorX Inference host | DeepSeek V4.1 Flash | $0.50 | $1.50 | $0.13 | $0.75 | 1M | yes |
| TensorX Inference host | MiniMax M3 | $0.40 | $2.00 | $0.10 | $0.80 | 1M | yes |
| TensorX Inference host | Qwen3.8 27B | $0.40 | $2.40 | $0.10 | $0.90 | 256K | yes |
| TensorX Inference host | Kimi K2.7 Code | $1.25 | $4.50 | $0.31 | $2.06 | 256K | yes |
| TensorX Inference host | GLM-5.2 | $1.50 | $4.50 | $0.38 | $2.25 | 1M | yes |
| TensorX Inference host | GLM-5.3 | $1.75 | $4.50 | $0.44 | $2.44 | 1M | yes |
| TensorX Inference host | DeepSeek V4 Pro 0813 | $2.00 | $4.00 | $0.50 | $2.50 | 1M | yes |
| TensorX Inference host | Qwen3.8 2.4T A95B | $2.50 | $6.00 | $0.63 | $3.38 | 256K | yes |
| TensorX Inference host | Kimi K3 | $3.00 | $15.00 | $0.75 | $6.00 | 1M | yes |
Per 1M tokens, standard (non-promo) list prices; short-context / peak tier where tiered. USD unless shown with € or CHF: those are as published and not converted. Click a header to sort. 11 models.
* Blended assumes 3 input tokens for every output token: (3×input + output) / 4. Your real mix will differ, so use the cost calculator with your own token counts for a real estimate.