27B
Parameters
Dense Transformer
AIVory Smart Inference · Qwen
Open WeightQwen 3.5 27B - built for coding agents.
Alibaba's latest small model. 27B parameters, tuned specifically for agentic coding workflows and function calling. Cheap enough to run in tight loops.
At a glance
Qwen 3.5 27B 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 3.5 27B: 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 3.5 27B 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-3.5-27b",
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-3.5-27b",
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-3.5-27b",
"messages": [{"role": "user", "content": "Ship it."}]
}'That's the whole migration. Get early access and try it →
Use cases
What Qwen 3.5 27B is great at.
-
IDE Coding Agents
Qwen 3.5 is specifically tuned for edit-test-fix loops. It generates code, reads compiler errors, and tries again. The 27B size means each iteration costs fractions of a cent, so your agent can loop 20 times on a bug without blowing budget.
-
Function Calling Chains
Reliable tool_choice and structured JSON output. Give it five functions and a user request, and it picks the right one, formats the arguments correctly, and chains calls when needed. Lower error rate on function calling than many 70B models.
-
Lightweight Assistants
When you don't need frontier reasoning but you do need fast, cheap responses with good format compliance. Internal tools, Slack bots, documentation Q&A. Qwen 3.5 at 27B costs less than half of a 70B model per token.
FAQ
Common questions.
What does Qwen 3.5 27B cost on AIVory?
One of the cheaper models in the catalogue. The 27B parameter count means low per-token cost. Check the live table above for current rates.
Qwen 3.5 27B vs Qwen 3 235B - when to upgrade?
3.5 27B is faster, cheaper, and better at code and tool use. 3 235B is stronger at complex reasoning, multi-step analysis, and long-form synthesis. Start with 3.5 for coding agents, move to 3 235B when quality isn't enough.
How does Qwen 3.5 compare to Mistral Small 24B for function calling?
Both are strong at structured output. Qwen 3.5 has the edge on coding tasks and multi-step chains. Mistral Small has better multilingual coverage for European languages. Both are in the same price range.
Can I use Qwen 3.5 for non-coding tasks?
Yes. The agentic coding tuning doesn't hurt general performance. It handles summarization, extraction, and chat fine. But if your workload is purely text with no code or tool use, Gemma 4 31B may be a better fit.
Start using Qwen 3.5 27B. Today.
Swap your base_url, keep shipping.
The router is live. Credits from $10, no subscription, no commitment.