24B
Parameters
Dense Transformer
AIVory Smart Inference · Mistral
Open WeightMistral Small 3.2 - vision + function calling.
The June 2025 update to Mistral Small. Now accepts image input alongside text, 128K context (4x the original), and the same structured output reliability.
At a glance
Mistral Small 3.2 24B via Smart Inference.
Routed to the cheapest live provider on every request. Same model, lower bill.
128K
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 3.2 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 3.2 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-3.2-24b-instruct-2506",
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-3.2-24b-instruct-2506",
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-3.2-24b-instruct-2506",
"messages": [{"role": "user", "content": "Ship it."}]
}'That's the whole migration. Get early access and try it →
Use cases
What Mistral Small 3.2 24B is great at.
-
Visual Document Understanding
Send a screenshot of an invoice, receipt, or form. Mistral Small 3.2 reads the image and returns structured data. Combine this with function calling to push extracted fields directly into your database or ERP.
-
Chart and Diagram Analysis
Feed it a bar chart, architecture diagram, or UI mockup alongside a text question. It describes what it sees and answers questions about the visual content. Cheaper than Maverick for simple image tasks.
-
128K Context Tool Chains
The 4x context upgrade from the original Mistral Small means you can include larger retrieval windows in RAG pipelines while still using function calling to act on the results.
FAQ
Common questions.
What does Mistral Small 3.2 cost on AIVory?
Priced similarly to the original Mistral Small 24B. Vision requests may cost slightly more depending on image size. Check the live table above for current rates.
What changed from Mistral Small 24B to 3.2?
Two things: image input support and a 128K context window (up from 32K). Everything else carries over, including strong function calling and structured output. The parameter count is still 24B.
Can it handle complex images like scanned documents?
It handles clean screenshots, charts, and diagrams well. For noisy scans or handwritten text, results are less reliable. Test with your actual document types before building a pipeline around it.
Mistral Small 3.2 vs Llama 4 Maverick for image tasks?
Maverick is a much larger model (400B MoE) and handles complex visual reasoning better. Mistral Small 3.2 is cheaper and faster for simple image tasks like form extraction and chart reading. Pick based on image complexity.
Start using Mistral Small 3.2 24B. Today.
Swap your base_url, keep shipping.
The router is live. Credits from $10, no subscription, no commitment.