24B
Parameters
Dense Transformer
AIVory Smart Inference · Mistral
Open WeightMistral Small 24B - structured output, nailed.
Mistral's 24B dense model from January 2025. Tuned for function calling and structured formats. When your pipeline needs valid JSON every time, this is a safe pick.
At a glance
Mistral Small 24B via Smart Inference.
Routed to the cheapest live provider on every request. Same model, lower bill.
32K
Context window
tokens per request
—
Cheapest input price
per 1M tokens via Smart Inference
—%
Cheaper
vs the priciest endpoint for the same model
Pricing
Mistral Small 24B: price comparison.
Our cheapest routed price vs what you'd pay elsewhere.
Prices in USD per 1M input tokens · updated live from the router
Quick start
Use Mistral Small 24B in three lines.
OpenAI-compatible. Swap your base_url, keep everything else.
from openai import OpenAI
client = OpenAI(
base_url="https://smart.aivory.net/v1",
api_key="sk-aivory-...",
)
resp = client.chat.completions.create(
model="mistral-small-24b-instruct-2501",
messages=[{"role": "user", "content": "Ship it."}],
)
print(resp.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://smart.aivory.net/v1",
apiKey: process.env.AIVORY_API_KEY,
});
const resp = await client.chat.completions.create({
model: "mistral-small-24b-instruct-2501",
messages: [{ role: "user", content: "Ship it." }],
});
console.log(resp.choices[0].message.content);curl https://smart.aivory.net/v1/chat/completions \
-H "Authorization: Bearer $AIVORY_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "mistral-small-24b-instruct-2501",
"messages": [{"role": "user", "content": "Ship it."}]
}'That's the whole migration. Get early access and try it →
Use cases
What Mistral Small 24B is great at.
-
Reliable Function Calling
Give Mistral Small a set of tool definitions and it picks the right one with correctly formatted arguments. Lower hallucination rate on function names and parameter types than most models in this size range.
-
JSON Schema Compliance
Define a JSON schema in your system prompt and Mistral Small follows it. Use it for data extraction pipelines where downstream systems reject malformed output. The structured output consistency saves debugging time.
-
Multilingual Customer Support
Handles French, German, Spanish, Italian, and Portuguese well. Mistral trained it with European language focus. Run a single model for EU customer support instead of maintaining per-language systems.
FAQ
Common questions.
What does Mistral Small 24B cost on AIVory?
In the same price range as other 24B-27B models. Cheaper than any 70B model but more expensive than Nemo 12B. Check the live table above for current rates.
Mistral Small 24B vs Mistral Small 3.2 - which one?
The 3.2 version adds image input and a 128K context window (vs 32K). If you need vision or long context, use 3.2. For text-only structured output with shorter inputs, the original is fine and may be slightly cheaper.
How does it compare to Qwen 3.5 27B for function calling?
Both are strong. Mistral Small has better European language support. Qwen 3.5 is better at coding and multi-step tool chains. If your tools are mostly API calls with simple schemas, either works.
Is 32K context enough?
For most structured output tasks, yes. Function calling requests are typically short. If you're processing long documents, consider Mistral Small 3.2 (128K) or Mistral Nemo (128K, 12B) instead.
Start using Mistral Small 24B. Today.
Swap your base_url, keep shipping.
The router is live. Credits from $10, no subscription, no commitment.