Romulus Hill

Build with affordable inference, or earn from your GPU @Scalattice

CEO @VerefaID @Robottik_ @Scalattice @WithTheMarket_

[Romulus Hill @romulushill]

@AgenticJonesy @BitcoinRhys We've got it affordably @Scalattice if you want to give it a go without using your RAM

[Romulus Hill @romulushill]

@TheDanielSto We've got insanely affordable open models @Scalattice if you want to get Qwen3.8-27b

[Romulus Hill @romulushill]

@yashmp2004 You can also get cheap inference if you are short on ram @Scalattice so you've got the ultimate Linux setup without shorting on inference capbailities.

[Romulus Hill @romulushill]

@0xSero We’ve got it hosted affordable on @Scalattice if you want to keep your own RAM free!

[Scalattice @Scalattice]

Qwen3-Coder 30B. OpenAI SDK. $5 credit, no card required.

OPENAI_BASE_URL=https://t.co/KhZHp5XodB
model=qwen-3-coder-30b-a3b

$0.067 / $0.267 per million.

[Scalattice @Scalattice]

We've decided to offer inference for Qwen-3-Coder-30B-A3B

We're charging $0.067 / $0.267 per million tokens.

Giving away free tokens to get started, no card required.

OPENAI_BASE_URL=https://t.co/cKcj9Z5cXg

Give it a try, let us know how it works out for you!

[Romulus Hill @romulushill]

Probably the most affordable and best way to get a coding model!

[Scalattice @Scalattice]

Programmers and production workloads, we've got affordable inference for you!

Join Scalattice and start with Free tokens today.

Qwen-3-coder-30b-a3b is available on Scalattice alongside our affordable open weights inference.

[Scalattice @Scalattice]

Programmers and production workloads, we've got affordable inference for you!

Join Scalattice and start with Free tokens today.

Qwen-3-coder-30b-a3b is available on Scalattice alongside our affordable open weights inference.

[Romulus Hill @romulushill]

@MikeBradleyAI We’re offering the ability to connect your GPU to our network @Scalattice to earn from inference of you are interested!

[Romulus Hill @romulushill]

@alexandre_ganso @MultiverseCompu @ArtificialAnlys @Scalattice It's important to note it's not currently open weights, but as and when release as such, we will be hosting European models.

[Romulus Hill @romulushill]

@alexandre_ganso @MultiverseCompu @ArtificialAnlys Hi Alexandre, as a rapidly growing inference network, we will be hosting it soon on @Scalattice I'm the founding engineer, and its our top priority to serve European Open Models to advance AI research in the EU!

[Romulus Hill @romulushill]

@cody_bunch @bckirsch We've got affordable hosted Ornith1.5 9b and 35b if you are interested in preserving your own ram!
@Scalattice

[Romulus Hill @romulushill]

@LMtxIoTAdvisor @loktar00 We could provision Qwen3.8 35B on our inference network @Scalattice if you would be interested!

We could do Qwen 4 35B and/or Qwen3.8 35B

I'm more than down to do that and discuss models you'd like to see!

[Romulus Hill @romulushill]

@goodpeace_PR We've got hosted Ornith1.5 on @Scalattice for both 9B and 35B if you are interested!

[Romulus Hill @romulushill]

@Blackwellboy @RealMatthewR We've got hosted Ornith1.5 for 9B and 35B @Scalattice if you want a fast way to try it out!

[Romulus Hill @romulushill]

@sebuzdugan @Scalattice Hi @sebuzdugan I'm a founding engineer at Scalattice, wondered if you'd like to discuss a custom billing plan so you can explore with some of our hosted models like Ornith1.5?

[Romulus Hill @romulushill]

Thanks mate! Please reach out with feedback if you want to experiment with a specific hosted model, and we will provision it to our fleet!

One of our main advantages of running our network, is we don't own any of the physical hardware, so we are able to offer much more affordable token rates on our models compared to rival inference companies.

So if you are looking for some of the most affordable offerings in the world, we've got them!

[Romulus Hill @romulushill]

Hi Chris, surprisingly as a new network we have a large volume of provider machines on our network all earning to-standard in line with our predictions.

We've got a good pool of developers too (who are paying for inference) which keeps it flowing!

As a new inference network we are working to provide a good balance of powerful models, which can still be run on consumer hardware from low to high end cards and machines, allowing us to compete with cloud providers like Salad, Vast ai, etc on medium to large models, whilst providing a very strong backbone of medium to low sized models.

It's an exciting business model, and so we are always keen to onboard more potential developers like yourself, to evaluate our networks capabilities etc.

Great to connect man!

[Romulus Hill @romulushill]

@Chris_Wozniczek @Scalattice or contact*

[Romulus Hill @romulushill]

@Chris_Wozniczek @Scalattice Yeah no worries man, that's exactly the kind of use case we are intending with our hosted offerings.

Don't hesitate to send me a message of contact our [email protected] if you want help getting started!