We use cookies. Cookies and browser storage keep you signed in and remember your preferences. With your permission we also measure how the site is used. You can any time. See our cookie policy.
Meta: Llama 3.3 70B Instruct
meta-llama/llama-3.3-70b-instructTry Meta: Llama 3.3 70B Instruct now
Runs on your balance at the prices above. New accounts start with free usage.
One API. Every model.
Session cost $0.00
Settings
Advanced parameters
Meta: Llama 3.3 70B Instruct at a glance
- Input price is 78% lower than the median of 434 Text models.
- Context window of 131,072 tokens, enough for long documents and multi-turn conversations.
- Supports tool (function) calling for agents and integrations.
- Supports structured JSON output with a schema.
- A typical request (1,000 input / 500 output tokens) costs about $0.000910.
About Meta: Llama 3.3 70B Instruct
Meta: Llama 3.3 70B Instruct is a text generation model from Meta, released on 6 December 2024. It belongs to the Llama 3.3 family, a multilingual large language model that Meta has both pretrained and instruction tuned at 70B parameters. The instruction tuned variant is text only, meaning it is designed for prompt driven tasks rather than multimodal input.
The model takes text in and returns text out, with a context window of 131,072 tokens and a maximum output of 16,384 tokens. Its knowledge cutoff is 31 December 2023, so anything after that date is outside what it learned during training. Because it is instruction tuned, it follows directions in a prompt reasonably directly, which makes it suitable for tasks where you describe what you want rather than supply examples.
Those capabilities point to a fairly wide set of practical uses. Long context work such as summarising or querying lengthy documents fits the 131,072 token window, and the 16,384 token output ceiling supports responses that run well beyond a short answer. Multilingual generation and chat style assistants are natural fits given the model's multilingual training. Instruction following also makes it a reasonable choice for extraction, rewriting, classification and similar transformation tasks where the input is text and the output is text.
You call it through api14's OpenAI-compatible API using the model id "meta-llama/llama-3.3-70b-instruct". The endpoint accepts the usual chat style request shape, and the model supports a broad set of parameters: frequency_penalty, logit_bias, logprobs, max_tokens, min_p, presence_penalty, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs and top_p. Tool calling and structured outputs are both available, so function calling and schema constrained responses are options rather than workarounds.
Two limits are worth keeping in mind when you choose this model. It accepts text only, so image, audio or other non-text inputs are out of scope, and its knowledge stops at the end of 2023, which matters for questions about recent events unless you supply the context yourself. If your task depends on those things, a model with the relevant input modality or a more recent cutoff would fit better.
How to call the Meta: Llama 3.3 70B Instruct API
- Create an API key in your dashboard.
- Install the official openai SDK for your language.
- Set baseURL to https://api14.com/api/v1 and model to "meta-llama/llama-3.3-70b-instruct".
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api14.com/api/v1",
apiKey: process.env.API14_API_KEY,
});
const result = await client.chat.completions.create({
"model": "meta-llama/llama-3.3-70b-instruct",
"messages": [
{
"role": "user",
"content": "Explain vector databases in two sentences."
}
]
});
console.log(result.choices[0].message.content);
Supported parameters
frequency_penaltylogit_biaslogprobsmax_tokensmin_ppresence_penaltyrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_pSpecifications
- Context length
- 131,072
- Max output
- 16,384
- Input
- text
- Output
- text
- Knowledge cutoff
- 2023-12-31
Requests on api14 · last 7 days
Meta: Llama 3.3 70B Instruct: frequently asked questions
- How much does the Meta: Llama 3.3 70B Instruct API cost?
- Meta: Llama 3.3 70B Instruct costs Input $0.35 / 1M tokens. It is charged from your USD balance only for successful requests; failed requests are free.
- What is the context window of Meta: Llama 3.3 70B Instruct?
- Meta: Llama 3.3 70B Instruct reads up to 131,072 tokens per request: prompt, history and documents together. It writes up to 16,384 output tokens per response.
- What can Meta: Llama 3.3 70B Instruct do?
- Meta: Llama 3.3 70B Instruct accepts text and returns text. It supports tool calling, structured JSON output.
- How do I call Meta: Llama 3.3 70B Instruct from my code?
- Point any OpenAI SDK at https://api14.com/api/v1 and set the model to "meta-llama/llama-3.3-70b-instruct". No other code changes. Copy a ready example from the code panel on this page.
- What are alternatives to Meta: Llama 3.3 70B Instruct?
- Comparable Text models at a similar price: Xiaomi: MiMo-V2.6-Flash, Qwen: Qwen3.5-9B, Google: Gemma 3 27B. Each runs through the same API key, so switching is one string.
Meta: Llama 3.3 70B Instruct community
New topicThis community has no topics yet. Open the first one about Meta: Llama 3.3 70B Instruct.