Everything about Llama-3.3-70B-Instruct

What you need to know

Released on December 6, 2024, Llama-3.3-70B-Instruct is one of the 7 Meta models tracked on this site. It is an open-weight model: you can host it yourself on infrastructure you control, and 43 API providers listed on models.dev also offer it. It has no declared reasoning mode: it answers directly, without an extra billed thinking step. That makes it more than a year old: successors may have shipped since.

We file it under open-weight models (68 models in this category on the site).

Benchmark results

Epoch AI publishes 2 results for Llama-3.3-70B-Instruct on our basket's benchmarks: too few to compute a reliable capability index.

• GPQA Diamond (PhD-level science questions): 47.4% — the best tracked model, Gemini 4 Argon, reaches 96.4%
• OTIS Mock AIME (competition mathematics problems): 5.1% — the best tracked model, GPT-5.5 Pro, reaches 100%

What it costs

Llama-3.3-70B-Instruct costs $0.60 per million input tokens and $0.76 per million output tokens (median of 43 providers, checked on October 22, 2026).

In practice, an agent task that reads 20,000 tokens and writes 2,000 costs about $0.014, or $13.52 for 1,000 such tasks.

That is 1.3 times cheaper than the median output price of open-weight models ($1).

To estimate your own budget, use the agent cost calculator.

Technical specifications

Its context window reaches 128,000 tokens, roughly 192 pages of text at 500 words per page: enough for a medium-length document. A single answer can run to 4,096 tokens.

It accepts as input: text. Its knowledge stops at December 2023: beyond that, it needs web search or your documents.

It can call tools (functions, APIs), which makes it a possible foundation for an AI agent.

The weights are published on Hugging Face (meta-llama/Llama-3.3-70B-Instruct). See also our list of open-source models.

What is it good for?

Given its category, Llama-3.3-70B-Instruct is mainly aimed at projects where you want control over your data or hosting, or want to adapt the model to your field.

Before adopting it, test it on your own examples: benchmarks measure standard tasks, not your job.

What could replace it?

In the same category, the models closest to Llama-3.3-70B-Instruct are:

• Trinity Large Thinking (Arcee AI, $0.80 per million output tokens)
• MiMo-V2.6-Pro (Xiaomi (MiMo), $0.87 per million output tokens)
• MiMo-V2.5-Pro (Xiaomi (MiMo), $0.87 per million output tokens)
• DeepSeek V4 Flash Vision Exp (DeepSeek, $0.65 per million output tokens)