AI Model Watch
Models / Meta

Llama 3.1 70B Instruct

GA

Open-weight, text-only, 70B dense, 128k context. License: Llama 3.1 Community License. Largely superseded by Llama 3.3 70B Instruct (same size, improved quality, Dec 2024) which Meta positions as the recommended 70B; not formally 'deprecated/retired' (open weights remain downloadable), so 'replacement' is advisory, not an enforced retirement. retires_on field repurposed here to point to the recommended successor id (no official retirement date exists - treat as null date). Per-token hosted price not separately captured this pass (null).

Llama 3.1 70B Instruct by Meta is tracked live in our LLM catalog, with a 128K (128.000-token) context window. It is generally available (GA), with a knowledge cutoff of Dec 2023. Its weights are publicly downloadable, so it can be self-hosted on your own infrastructure.

Last verified: 13 Aug 2026 · sourced from official provider documentation

Provider
Meta
Status
GA
Input price
—
Output price
—
Cached input
—
Blended price
—
Context window
128.000 tokens (128K)
Max output
—
Modality
text
Open weights
Yes — self-hostable
Knowledge cutoff
2023-12
Released
23 Jul 2024
API string
meta-llama/Llama-3.1-70B-Instruct

Source: Meta official documentation ↗

See also

Explore more

All Meta models→

Prices, context windows and lifecycle status for every Meta model we track.

Compare models head-to-head→

Side-by-side price, context and status for the models developers weigh against each other.

FAQ

Llama 3.1 70B Instruct — questions & answers

What is the context window of Llama 3.1 70B Instruct?

Llama 3.1 70B Instruct has a 128.000-token context window (128K).

Is Llama 3.1 70B Instruct deprecated?

No — Llama 3.1 70B Instruct is generally available (GA) and not currently scheduled for deprecation or retirement in our tracker.

Does Llama 3.1 70B Instruct support image or vision input?

Our catalog lists Llama 3.1 70B Instruct's modalities as text; image/vision input is not among them.

Can I self-host Llama 3.1 70B Instruct?

Yes — Llama 3.1 70B Instruct is an open-weight model: its weights are publicly downloadable, so you can run it on your own infrastructure (subject to its license) instead of only calling Meta's API.

What is the API model string for Llama 3.1 70B Instruct?

The API model identifier for Llama 3.1 70B Instruct is "meta-llama/Llama-3.1-70B-Instruct" when calling Meta's API.

What is Llama 3.1 70B Instruct's knowledge cutoff?

Llama 3.1 70B Instruct's training knowledge cutoff is Dec 2023.