70B
Parameters
Dense Transformer
AIVory Smart Inference · Llama
Open WeightLlama 3.3 70B - the open-weight workhorse.
The most widely deployed open-weight model. Hosted by nearly every inference provider, which means more competition on price. 128K context, reliable output.
At a glance
Llama 3.3 70B 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
Llama 3.3 70B: 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 Llama 3.3 70B 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="llama-3.3-70b",
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: "llama-3.3-70b",
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": "llama-3.3-70b",
"messages": [{"role": "user", "content": "Ship it."}]
}'That's the whole migration. Get early access and try it →
Use cases
What Llama 3.3 70B is great at.
-
General-Purpose Assistant Backend
Llama 3.3 70B is the default choice when you need a solid model and don't have a specific reason to pick something else. Summarization, Q&A, text rewriting, data extraction. It handles all of them consistently.
-
Coding Assistants
Strong on Python, JavaScript, Java, and TypeScript generation. Use it as the backbone for IDE plugins, code review bots, or test generators. Most providers serve it with sub-second first-token latency.
-
High-Availability Workloads
More providers host Llama 3.3 than any other open model. That means the router always has fallback options. If your top priority is uptime over peak quality, 3.3 is the safest pick.
FAQ
Common questions.
How cheap is Llama 3.3 70B on AIVory?
Because so many providers compete on Llama 3.3, it's often the cheapest 70B-class model available. The priciest endpoint can cost 3x the cheapest. Smart Inference always picks the lowest. Check the live table above.
Is Llama 3.3 still worth using with Llama 4 available?
Yes. Llama 3.3 is faster and cheaper per token than Maverick or Scout. For tasks that don't need multimodal input or frontier-level reasoning, 3.3 gives you better cost efficiency.
How many providers serve Llama 3.3?
More than any other model in the catalogue. The exact count changes, but the router typically evaluates 5+ candidates per request. Check the X-SI-Candidates header on your responses to see the current count.
Can I use Llama 3.3 for commercial products?
Yes. Meta's Llama 3.3 Community License allows commercial use. There are no royalties or usage caps for applications under 700M monthly active users.
Llama 3.3 70B vs Qwen 2.5 72B?
Similar size, similar cost. Llama 3.3 has broader provider coverage and slightly better English benchmarks. Qwen 2.5 is stronger on multilingual tasks and math. Pick based on your workload.
Start using Llama 3.3 70B. Today.
Swap your base_url, keep shipping.
The router is live. Credits from $10, no subscription, no commitment.