# Llama 3.1 405B is published **URL:** \ **Category:** Discussions\ **Created:** [July 23, 2024, 4:09pm UTC](https://forum.cursor.com/t/llama-3-1-405b-is-published/6684 "2024-07-23T16:09:53Z")\ **Posts on this page:** 7\ **Page:** 1
**Author:** ![Orpheus](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/orpheus/32/9_2.png) [@Orpheus](https://forum.cursor.com/u/Orpheus)\ **Post date:** [July 23, 2024, 4:09pm UTC](https://forum.cursor.com/t/llama-3-1-405b-is-published/6684/1 "2024-07-23T16:09:53Z")
> **[Industry Leading, Open-Source AI | Llama](https://www.llama.com/)** > > Discover Llama 4's class-leading AI models, Scout and Maverick. Experience top performance, multimodality, low costs, and unparalleled efficiency. > **[The Llama 3 Herd of Models | Research - AI at Meta](https://ai.meta.com/research/publications/the-llama-3-herd-of-models/)** > > Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a... ### Key Facts about Llama 3.1 405B (generated by gpt4o and i’m not sure if 100% accurate) - **Parameters** : 405 billion - **Training Tokens** : 15.6 trillion - **Compute Used** : Approximately 3.8e25 floating-point operations (FLOP) - **Training Duration** : Approximately 72 days - **GPU Usage** : Trained using a cluster of 16,000 H100 GPUs - **GPU Performance** : Each GPU performed 380 teraFLOP per second - **Compute Intensity** : Around twice the compute of GPT-4 - **Status** : Most compute-intensive open-weight model to date - **Multimodal Capabilities** : Supports integration of image, video, and speech inputs using a compositional approach - **Release and Licensing** : Publicly released under Llama 3 Community License ### Benchmark Performance Comparison | **Benchmark** | **GPT-4o** | **Llama 3.1-405B** | | --- | --- | --- | | **BoolQ** | 0.905 | 0.921 | | **GSM8K** | 0.942 | 0.968 | | **HellaSwag** | 0.891 | 0.920 | | **HumanEval** | 0.921 | 0.854 | | **MMLU\_Humanities** | 0.802 | 0.818 | | **MMLU\_Other** | 0.872 | 0.875 | | **MMLU\_Social\_Sciences** | 0.913 | 0.898 | | **MMLU\_STEM** | 0.696 | 0.831 | | **OpenBookQA** | 0.882 | 0.908 | | **PIQA** | 0.844 | 0.874 | | **Social IQA** | 0.790 | 0.797 | | **TruthfulQA\_MC1** | 0.825 | 0.800 | | **Winogrande** | 0.822 | 0.867 | ---
**Author:** ![gpickett00](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/gpickett00/32/30563_2.png) [@gpickett00](https://forum.cursor.com/u/gpickett00)\ **Post date:** [July 23, 2024, 5:02pm UTC](https://forum.cursor.com/t/llama-3-1-405b-is-published/6684/2 "2024-07-23T17:02:39Z")
Is there a way to use it inside of Cursor yet? ---
**Author:** ![deanrie](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/deanrie/32/1263_2.png) [@deanrie](https://forum.cursor.com/u/deanrie)\ **Post date:** [July 24, 2024, 3:55am UTC](https://forum.cursor.com/t/llama-3-1-405b-is-published/6684/3 "2024-07-24T03:55:32Z")
You can use the OpenRouter [Meta: Llama 3.1 405B Instruct by meta-llama – Run with a standardized API | OpenRouter](https://openrouter.ai/models/meta-llama/llama-3.1-405b-instruct) ---
**Author:** ![Kirai](https://avatars.discourse-cdn.com/v4/letter/k/fbc32d/32.png) [@Kirai](https://forum.cursor.com/u/Kirai)\ **Post date:** [July 25, 2024, 8:25am UTC](https://forum.cursor.com/t/llama-3-1-405b-is-published/6684/4 "2024-07-25T08:25:49Z")
> [@Orpheus](#): > > **Multimodal Capabilities** : Supports integration of image, video, and speech inputs using a compositional approach I think this isn’t in the released model. [https://www.perplexity.ai/search/recently-released-llama-3-1-40-mXuB3I9BTuqCSZtlPFeBjg](https://www.perplexity.ai/search/recently-released-llama-3-1-40-mXuB3I9BTuqCSZtlPFeBjg) ---
**Author:** ![ssmits](https://avatars.discourse-cdn.com/v4/letter/s/aca169/32.png) [@ssmits](https://forum.cursor.com/u/ssmits)\ **Post date:** [July 25, 2024, 8:41pm UTC](https://forum.cursor.com/t/llama-3-1-405b-is-published/6684/5 "2024-07-25T20:41:46Z")
Openrouter will work however it is still not possible to easily switch between API calls and calls linked to the Pro plan (the 500). @truell20 ---
**Author:** ![Orpheus](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/orpheus/32/9_2.png) [@Orpheus](https://forum.cursor.com/u/Orpheus)\ **Post date:** [July 29, 2024, 1:13pm UTC](https://forum.cursor.com/t/llama-3-1-405b-is-published/6684/6 "2024-07-29T13:13:15Z")
Curious if anyone that used for complex tasks in Cursor can report about its performance. ---
**Author:** ![system](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/system/32/99069_2.png) [@system](https://forum.cursor.com/u/system)\ **Post date:** [June 18, 2026, 10:13am UTC](https://forum.cursor.com/t/llama-3-1-405b-is-published/6684/7 "2026-06-18T10:13:59Z")