# Llama Guard 4 12B vs Mistral Small (24B) Instruct 25.01

On provider list prices, Llama Guard 4 12B costs $0.20 per million input tokens against $0.10 for Mistral Small (24B) Instruct 25.01: effectively level. Output is $0.20 against $0.30 (1.5x). On Allocate both bill at list plus the 7% transaction fee.

## Specifications

| | Llama Guard 4 12B | Mistral Small (24B) Instruct 25.01 |
| --- | --- | --- |
| Lab | Meta | mistralai |
| Access | Open weights | Open weights |
| Context window | 1M tokens | 32K tokens |
| List price, input | $0.20 / M tokens | $0.10 / M tokens |
| List price, output | $0.20 / M tokens | $0.30 / M tokens |
| Cached input | n/a | n/a |
| License | Llama community | Apache 2.0 |
| Fine-tunable | Yes | Yes |

Specifications and provider list prices from the Allocate catalog, checked 2026-07-08. Billed price is list plus the 7% transaction fee.

## What the numbers say

Take 1,000,000 requests a month at 1,200 input and 350 output tokens each. That workload costs $225 a month on Mistral Small (24B) Instruct 25.01 and $310 on Llama Guard 4 12B at list: a gap of $85, or 1.4x.

Llama Guard 4 12B reads 1M tokens per request against 32K for Mistral Small (24B) Instruct 25.01, 32.0x the window. That decides which one can take whole documents without splitting them.

## Choose Llama Guard 4 12B for

- The longer context window (1M vs 32K tokens)

## Choose Mistral Small (24B) Instruct 25.01 for

- The lower list price ($0.10 in / $0.30 out per M tokens)
- Fine-tuning under a permissive license (Apache 2.0)

## Common questions

### Which is cheaper, Llama Guard 4 12B or Mistral Small (24B) Instruct 25.01?

Mistral Small (24B) Instruct 25.01, on this workload shape. At list prices it is $0.10/$0.30 per million tokens in and out against $0.20/$0.20 for Llama Guard 4 12B. Billed on Allocate: $0.11/$0.32 against $0.21/$0.21, list plus 7%.

### Which has the bigger context window?

Llama Guard 4 12B: 1,048,576 tokens (1M) against 32,768 (32K) for Mistral Small (24B) Instruct 25.01.

### Can I fine-tune Llama Guard 4 12B or Mistral Small (24B) Instruct 25.01?

Both publish open weights (Llama Guard 4 12B: Llama community; Mistral Small (24B) Instruct 25.01: Apache 2.0), so both can be fine-tuned. On Allocate the trained weights stay inside your boundary and belong to you.

---

[HTML page](https://allocate.network/compare/meta-llama-guard-4-12b-vs-mistral-mistral-small-24b-instruct-2501) · [Llama Guard 4 12B](https://allocate.network/models/meta-llama-guard-4-12b.md) · [Mistral Small (24B) Instruct 25.01](https://allocate.network/models/mistral-mistral-small-24b-instruct-2501.md) · [Machine-readable catalog](https://allocate.network/catalog.json)