# Run Mistral 7B v0.3

A fine-tuning base with no license strings attached. Ready in about 90 seconds on an RTX 4090, from $0.45/hr. No install, no code.

## What it does

Mistral 7B is a similar size and capability to Llama 3.1 8B, and trains just as easily on one card.

The reason to pick it is the license. It is Apache 2.0, so the model you train is yours to ship in a product with no naming requirements, no user thresholds, and no obligations.

If you are fine-tuning something that will end up inside a product you sell, start here rather than with Llama.

## Your setup

We run Mistral 7B v0.3 on an RTX 4090 with 24GB of memory, at $0.45 per hour, billed by the minute. It takes about 2 seconds per example.

Same reasoning as Llama 3.1 8B. LoRA training on a 7 billion parameter model needs about 15GB with a reasonable batch size, and a 24GB card covers that with headroom for longer training examples. Going larger buys you a bigger batch size and a modestly shorter run, which does not pay for itself at this scale.

## What it costs

1,000 training examples takes about 35 minutes of GPU time, which is about $0.26.

## Two ways to run it

- **Open the app.** JupyterLab with the training stack ready to go.
- **Run a batch job.** Spreadsheet of examples in, trained adapter out.

## Model facts

| Field | Value |
|---|---|
| Parameters | 7.3B |
| License | Apache 2.0 |
| Memory required | 15GB minimum for LoRA |
| Base model | Trained from scratch by Mistral AI |
| Publisher | Mistral AI |
| Recommended hardware | RTX 4090, 24GB |
| Category | [Fine-tuning](https://gpuvault.io/working-on/fine-tuning/) |

## Questions people ask

### Do I need to install anything to use Mistral 7B v0.3?

No. We start a machine with Mistral 7B v0.3 already loaded and hand you a link. Everything runs in your browser, there is nothing to download, and nothing is left on your computer afterwards. It works the same on a Mac, a Windows laptop, or a Chromebook. A workspace is usually ready in about 90 seconds.

### How much does it cost to run 1,000 training examples?

About $0.26. Mistral 7B v0.3 takes about 2 seconds per example on the RTX 4090 we recommend, so 1,000 training examples is roughly 35 minutes of GPU time at $0.45 per hour. Billing is by the minute, and a batch job shuts the GPU off the moment the last item finishes. The calculator above works this out for your own numbers.

### Can I run Mistral 7B v0.3 on a cheaper card?

Sometimes, and the calculator will not always make it look worth it. Same reasoning as Llama 3.1 8B. LoRA training on a 7 billion parameter model needs about 15GB with a reasonable batch size, and a 24GB card covers that with headroom for longer training examples. Going larger buys you a bigger batch size and a modestly shorter run, which does not pay for itself at this scale. If you want to try a different card anyway, the advanced catalog lets you pick one and shows the estimated time before you commit.

### Can I use Mistral 7B v0.3 commercially?

Yes. Mistral 7B v0.3 is released under the Apache 2.0 license, which permits commercial use with no conditions worth worrying about. You can use the output in paid client work and in products you sell.

### What happens to my files after I stop?

Anything you save into your workspace folder stays in your account and is there when you come back. Everything else is destroyed: the machine is terminated and its disk is wiped. Batch job inputs and outputs are deleted from our storage 24 hours after you download them. We do not read your files or use them to train anything.

### What if it fails partway through?

You are not charged for work that did not complete, and failed sessions are refunded automatically. Batch jobs write each output as it finishes rather than at the end, so a failure at item 380 of 400 leaves you 379 usable files and a message saying what broke. Restarting picks up where it stopped rather than redoing the work.

---

*Source: https://gpuvault.io/model/mistral-7b-v0-3/ — GPUVault, GPU rental by the minute.*