72B
Parameters
Dense Transformer
AIVory Smart Inference · Qwen
Open WeightQwen 2.5 72B - multilingual from day one.
Alibaba Cloud's 72B dense transformer. Trained on one of the broadest multilingual corpora in open-weight models. 128K context, strong on math and code.
At a glance
Qwen 2.5 72B 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
Qwen 2.5 72B: 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 Qwen 2.5 72B 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="qwen-2.5-72b",
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: "qwen-2.5-72b",
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": "qwen-2.5-72b",
"messages": [{"role": "user", "content": "Ship it."}]
}'That's the whole migration. Get early access and try it →
Use cases
What Qwen 2.5 72B is great at.
-
CJK and Multilingual Applications
Qwen 2.5 handles Chinese, Japanese, Korean, Arabic, and European languages without quality dropoff. Run a single model for global customer support instead of maintaining separate models per region.
-
Financial and Mathematical Modeling
Top scores on GSM8K and MATH among 72B-class models. Use it for numerical analysis, revenue forecasting, or any task that requires precise step-by-step arithmetic over large datasets.
-
Code Generation Across Languages
Competitive on HumanEval and MBPP across Python, Java, TypeScript, and Go. Write functions, generate tests, or refactor existing code. Cheaper per token than Llama 4 Maverick for tasks that don't need 400B-level reasoning.
FAQ
Common questions.
How much does Qwen 2.5 72B cost on AIVory?
Qwen 2.5 is priced like other 70B-class models. The exact rate depends on provider availability. Check the live table above. Typical input costs are under $0.30 per million tokens.
Qwen 2.5 72B vs Llama 3.3 70B - which should I pick?
Qwen 2.5 has better multilingual coverage, especially for CJK languages. Llama 3.3 is more widely deployed and has slightly broader English-language benchmarks. For multilingual workloads, Qwen wins. For English-only, try both.
Does Qwen 2.5 support function calling?
Yes. Qwen 2.5 supports the OpenAI-compatible tools/tool_choice format. The router only selects providers that have function calling enabled when your request includes tools.
What are the commercial terms for Qwen 2.5?
Alibaba released Qwen 2.5 under the Qwen License, which allows commercial use. No royalties, no usage caps. You can also self-host it via the GPU Marketplace.
Start using Qwen 2.5 72B. Today.
Swap your base_url, keep shipping.
The router is live. Credits from $10, no subscription, no commitment.