---
title: "Llama 4 Behemoth 2T Benchmarks - RankedAGI"
description: "Detailed benchmark and metadata record for Llama 4 Behemoth."
source: "https://rankedagi.com/models/llama-4-behemoth-preview"
---

# Llama 4 Behemoth

| Field | Value |
| --- | --- |
| Organization | Meta |
| License | Open Source |
| Size | 2T |
| Version | Preview |
| Released | 2025-04-05 |
| Knowledge cutoff | 2024-08-01 |
| Last updated | 2026-04-11T19:49:56.189Z |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 48.1% | RankedAGI Coding Score |  |
| RankedAGI Agentic | Agents | 39.3% | RankedAGI Agentic Score |  |
| LiveCodeBench v5 | coding | 49.0% | LiveCodeBench - Evaluation of Large Language Models for Code |  |
| RankedAGI Reasoning | reasoning | 46.6% | RankedAGI Reasoning Score |  |
| GPQA Diamond | reasoning | 73.7% | Generalized Prefix Question Answering Score (Reasoning) PhD Level Reasoning |  |
| MMLU Pro | knowledge | 82.2% | Graduate-level visual problem-solving |  |
| MMLU | knowledge | 85.8% | Image Reasoning |  |
| MMMU | imaging | 76.1% | Multimodal Understanding College-level visual problem-solving |  |
| RankedAGI Math | math | 42.4% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 46.2% | Overall RankedAGI score |  |

## Model Links

| Title | URL |
| --- | --- |
| Release | https://ai.meta.com/blog/llama-4-multimodal-intelligence/ |