TensorX

Verified

TensorX is an Ireland-based provider in the GPU cloud category with 15 listed models. 14 of 15 models publish a cached-input rate.

EU-owned EU hosting GPU cloud

Visit website API docs

At a glance

Record

Ownership: TensorX Ltd (Company No. 796387, formerly Tensorix)
Headquarters: Dublin, Ireland
Hosting region: EU (Dublin, Helsinki, Paris)
Founded: Not published
Website: tensorx.ai

Summary

Scope

TensorX is an Ireland-based provider in the GPU cloud category with 15 listed models. 14 of 15 models publish a cached-input rate.

Own NVIDIA B300 clusters; dedicated inference available; 42+ models claimed, 15 on public price list.

Source: tensorx.ai ↗ · Updated Nov 4, 2026

API access

OpenAI-compatible

Access

API

KYC

Not published

Compliance

ISO 27001-readyGDPR Article 44 compliantDPA

Key models

15 listed

DeepSeek V4.1 Flash 1017 1M context · per 1M tokens Multimodal $0.5 / $1.5

GLM-5.3 1M context · per 1M tokens Inference $1.75 / $4.5

Kimi K3 1M context · per 1M tokens Multimodal $3 / $15

MiniMax M3.1 Flash 1M context · per 1M tokens Multimodal $0.3 / $1.2

GLM-5.3 Flash 1M context · per 1M tokens Multimodal $0.2 / $0.5

Qwen3.8 Flash-Next 256K context · per 1M tokens Multimodal $0.2 / $0.5

Qwen3.8 2.4T A95B 256K context · per 1M tokens Inference $2.5 / $6

DeepSeek V4.1 Flash 1M context · per 1M tokens Multimodal $0.5 / $1.5

DeepSeek V4 Pro 0813 1M context · per 1M tokens Inference $2 / $4

DeepSeek V4 Flash 0731 1M context · per 1M tokens Inference $0.25 / $0.3

Qwen3.8 27B 256K context · per 1M tokens Multimodal $0.4 / $2.4

GLM-5.2 1M context · per 1M tokens Inference $1.5 / $4.5

MiniMax M3 1M context · per 1M tokens Multimodal $0.4 / $2

Kimi K2.7 Code 256K context · per 1M tokens Multimodal $1.25 / $4.5

Qwen3 Embedding 8B 40K context · per 1M tokens Embedding $0.02 / $0

Pricing

List rates · USD

Model Type Input price Caching price Output price

DeepSeek V4.1 Flash 10171M context · per 1M tokens Multimodal $0.5 $0.13 $1.5

GLM-5.31M context · per 1M tokens Inference $1.75 $0.44 $4.5

Kimi K31M context · per 1M tokens Multimodal $3 $0.75 $15

MiniMax M3.1 Flash1M context · per 1M tokens Multimodal $0.3 $0.08 $1.2

GLM-5.3 Flash1M context · per 1M tokens Multimodal $0.2 $0.05 $0.5

Qwen3.8 Flash-Next256K context · per 1M tokens Multimodal $0.2 $0.05 $0.5

Qwen3.8 2.4T A95B256K context · per 1M tokens Inference $2.5 $0.63 $6

DeepSeek V4.1 Flash1M context · per 1M tokens Multimodal $0.5 $0.13 $1.5

DeepSeek V4 Pro 08131M context · per 1M tokens Inference $2 $0.5 $4

DeepSeek V4 Flash 07311M context · per 1M tokens Inference $0.25 $0.06 $0.3

Qwen3.8 27B256K context · per 1M tokens Multimodal $0.4 $0.1 $2.4

GLM-5.21M context · per 1M tokens Inference $1.5 $0.38 $4.5

MiniMax M31M context · per 1M tokens Multimodal $0.4 $0.1 $2

Kimi K2.7 Code256K context · per 1M tokens Multimodal $1.25 $0.31 $4.5

Qwen3 Embedding 8B40K context · per 1M tokens Embedding $0.02 - $0

List rates. Cached input shown where published; "-" means the provider does not publish a cache rate. Units follow the source rate card.

Infrastructure

Hosting

RegionEU

CitiesDublin, Helsinki, Paris

Provider groupGPU cloud

ZDRyes (default, every request)

DPAAvailable

Cached input policyCache read billed at the cached rate.

Access

API & payments

API shapeOpenAI-compatible

AccessAPI

KYCNot published

Cached input policyCache read billed at the cached rate.

Compliance

Source-backed

DPAAvailable

ZDRyes (default, every request)

RetentionZDR: yes (default, every request). DPA published.

Certifications

ISO 27001-readyGDPR Article 44 compliantDPA

Benchmarks

Coming soon

Coming soon.

Performance

Coming soon

Latency

Time to first token

Median and p95 response times for short and long prompts, measured from repeatable probes rather than vendor claims.

Coming soon

Throughput

Tokens per second

Streaming output rate and request throughput under steady load, with concurrency and context length recorded.

Coming soon

Efficiency

Cost per workload

Blended cost for a fixed workload, including cache reads and output tokens, so speed and price can be compared together.

Coming soon

Independent performance probes are being added. Vendor-reported numbers are not mixed into this section.