An alternative to Together AI.
French roots. European GPUs.

01 / French roots, European infrastructure

Choose a French partner with European inference infrastructure.

Blade is Shadow’s AI infrastructure brand. Shadow is a French company registered in Paris, identified in our service terms. Blade’s own inference GPUs are operated by Shadow in Europe. That combines a French cloud partner with a concrete European hosting option for Blade-hosted models.

The team brings ten years of cloud experience to the service. Running GPU workloads at scale means managing capacity, availability and responsiveness as everyday engineering work. You can read more about the cloud experience behind Blade.

Years of cloud experience

10

Instances per year

16M

GPUs in the fleet

12,000+

These figures describe Shadow’s cloud operations, the experience behind Blade.

Together AI’s serverless documentation says its serving region is not selectable; it points teams with regional requirements to dedicated endpoints. For Blade-hosted models, European GPU infrastructure is part of the offering you can evaluate with Models.

Company origin and processing location answer different questions. External model providers and optional tools can follow separate paths; our hosting and data handling guide explains what is documented. Assess the selected services and contractual terms for your project rather than treating a European GPU location as a guarantee about every data flow.

02 / Developer experience

Keep your SDK and focus on the application.

Models runs the inference infrastructure for you. Choose a model available to your account, configure your Blade key and base URL, and send a supported request. You do not need to deploy model weights or choose a GPU to use the catalogue.

That makes evaluation a contained piece of work: keep your application’s client, test representative prompts and compare the results. The OpenAI SDK setup guide covers the setup. Validate streaming, tool calls and the parameters you use before moving production traffic.

Together also provides hosted inference, usage billing and a path to dedicated capacity through its serverless offering. The reason to evaluate Blade is the combination of its integration, tools and European infrastructure, not an assumption that Together requires you to manage servers.

03 / More useful workflows

Turn a model response into a researched answer.

For an assistant that needs current information, Flash by Staan brings web search into the workflow. With an enabled account and a supported model, Blade handles the built-in search loop and returns source annotations. Executed searches are billed separately.

Blade’s API also offers web fetch and code execution, subject to the relevant access and model requirements. Evaluate the search integration with real questions: whether the sources are useful, whether the answer is grounded and how much the complete request costs. These are product-level decisions that a model benchmark alone cannot settle.

04 / Cost per useful result

Give flexible jobs a timing advantage.

Models has no subscription or minimum spend, with published model rates. Eligible models use dynamic pricing, so document batches, embeddings and evaluations that can wait can be scheduled around lower rates. The dynamic pricing guide explains how the applicable rate is set.

Together also offers usage pricing and batch discounts, as documented in its pricing overview. Compare your actual input and output volumes, tool usage and quality target. Blade’s opportunity is to use the right model at an appropriate time; a From rate alone is not a total-cost comparison.

Blade and Together AI at a glance.

Scroll sideways to compare both providers →