When was Llama-3.3-70B-Instruct released, and is it still maintained?

Llama-3.3-70B-Instruct was released on 2024-12-06. Its catalog status is stable. Check the model page for deprecation notices before you build a long-lived integration on it.

How does Llama-3.3-70B-Instruct score on benchmarks?

Recorded results for Llama-3.3-70B-Instruct: aa_intelligence_index 7.9; arena_text 65.8. Benchmark positions are a coarse filter only — run your own prompts against the candidate models before you commit, because ranking shifts with task shape and prompt style.

What is the max output length for Llama-3.3-70B-Instruct?

Llama-3.3-70B-Instruct returns up to 16,384 output tokens per request, within a 128,000-token context window. Cap the response with the max_tokens parameter so a single call cannot run past your budget.

How do you get a cache hit on Llama-3.3-70B-Instruct?

The Llama-3.3-70B-Instruct record lists prompt caching as a supported feature but publishes no cache-read rate, so cached-prefix billing is not part of the current pricing. Treat repeated prefixes as fresh input when you size the workload, and confirm the rate with the provider before designing for a cache saving.

Can you call Llama-3.3-70B-Instruct with the OpenAI SDK you already use?

Yes. Llama-3.3-70B-Instruct is reachable through an OpenAI-compatible endpoint, so an existing OpenAI SDK client works after pointing the base URL at MixRoute and setting the model parameter to llama-3.3-70b-instruct . Keep your current request and response handling.

Model detail

Llama-3.3-70B-Instruct

Provided by Meta

Pay-as-you-go

Meta's seventy-billion-parameter instruction-tuned language model supports multilingual dialogue and general text-generation tasks.

Model specs

Model ID: llama-3.3-70b-instruct

Context length: 128K
Max output: 16.384K
I/O modalities
Released: 2024-12

API endpoints

One MixRoute gateway, OpenAI-compatible

OpenAI-compatible /v1/chat/completions POST

Pricing

Flat rate, unit: /1M Tokens

Input

$0.7100 /1M Tokens

Completion

$0.7100 /1M Tokens

Selection summary

Quickly judge whether it fits your workload

What is this model good for?

Meta's seventy-billion-parameter instruction-tuned language model supports multilingual

Meta's seventy-billion-parameter instruction-tuned language model supports multilingual dialogue and general text-generation tasks.

What should you check before using it?

128,000 context, 16,384 output tokens

Model type: text. Verify supported features and quotas with the provider documentation. Knowledge cutoff: 2023-12. Released 2024-12-06.

Why use it through MixRoute?

Swap the base URL, keep the Llama-3.3-70B-Instruct client

MixRoute serves Llama-3.3-70B-Instruct on the OpenAI-compatible endpoint /v1/chat/completions; point your existing client at https://api.mixroute.ai/v1 with the model ID llama-3.3-70b-instruct and re-run your integration tests.

Capabilities & compatibility

Unconfirmed capabilities stay "Unknown," no guesses

Tool calling

Llama-3.3-70B-Instruct supports function and tool calls.

EX Send a tools definition; the response returns tool_calls to execute and feed results back.

Structured output

Llama-3.3-70B-Instruct can produce JSON-formatted output on demand.

EX response_format: { type: "json_schema" } forces valid JSON output.

Streaming

Llama-3.3-70B-Instruct supports streamed responses.

EX stream: true streams tokens incrementally, so chat UIs render as they arrive.

Prompt caching

Llama-3.3-70B-Instruct supports prompt caching under supported routing conditions.

EX Keep the system prompt and tool definitions at the start of the request; repeated prefixes are automatically billed at the cache-read rate.

OpenAI SDK

Call Llama-3.3-70B-Instruct through MixRoute's OpenAI-compatible endpoint.

EX Change base_url to https://api.mixroute.ai/v1; everything else follows the OpenAI style.

Token cost estimator

Live estimate from this page's pricing, not an actual bill

Input tokens Max per request 128K

Output tokens Max per request 16.384K

Requests / month

Cache hit rate –

This model has no cache-read price; caching is not counted

FAQ

Set max_tokens on Llama-3.3-70B-Instruct to the longest response your task actually needs, not the 16,384-token ceiling. Keeping the cap tight reduces worst-case spend and shortens tail latency. Raise it only for tasks that demonstrably need longer output.

Get started

01

Create an API key

Generate a key in the Console; one key calls every model.
02

Set the base URL

Fully OpenAI SDK-compatible; just swap the URL to connect.

https://api.mixroute.ai/v1
03

Specify the model ID

In the request's model field, enter:

llama-3.3-70b-instruct

Create an API key

New users get $5 in free credits

Test this model and alternate routes with the same request format

Start from a real workload, then let quality, total cost, and failure conditions decide whether to send production traffic.

Create an API key View integration docs