---
title: "Llama 3.3 70B Benchmarks - RankedAGI"
description: "Detailed benchmark and metadata record for Llama 3.3."
source: "https://rankedagi.com/models/llama-3.3-70b"
---

# Llama 3.3

| Field | Value |
| --- | --- |
| Organization | Meta |
| License | Open Source |
| Size | 70B |
| Version | Latest |
| Released | 2024-12-06 |
| Knowledge cutoff | 2023-12-01 |
| Input cost per million tokens | $0.1 |
| Output cost per million tokens | $0.4 |
| Last updated | 2025-04-16 22:14:46.146909+00 |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 35.2% | RankedAGI Coding Score |  |
| RankedAGI Agentic | Agents | 30.4% | RankedAGI Agentic Score |  |
| ChatArena (LMSYS) | coding | 1232 | ChatArena (LMSYS) Coding ELO Score with Style Control |  |
| LiveBench Coding Score (old) | coding | 36.6% | LiveBench Coding Score old version |  |
| RankedAGI Reasoning | reasoning | 37.4% | RankedAGI Reasoning Score |  |
| GPQA Diamond | reasoning | 50.5% | Generalized Prefix Question Answering Score (Reasoning) PhD Level Reasoning |  |
| Text Arena | general | 1257 | ChatArena (LMSYS) ELO Score | https://arena.ai/leaderboard/text |
| HumanEval | coding | 88.4% | HumanEval coding benchmark |  |
| HumanEval+ | coding | 87.6% | HumanEval+ benchmark |  |
| NYT Connections Extended | reasoning | 15.1% | NYT Connections Extended Version | https://github.com/lechmazur/nyt-connections/ |
| MMLU Pro | knowledge | 68.9% | Graduate-level visual problem-solving |  |
| MMLU | knowledge | 86.0% | Image Reasoning |  |
| Hughes Hallucination Evaluation | safety | 4.0% | Computed using Vectara's Hallucination Evaluation Model, also known as HHEM. This evaluates how often an LLM introduces hallucinations when summarizing a document. | https://github.com/vectara/hallucination-leaderboard |
| AidanBench | general | 779 | AidanBench Score |  |
| LiveBench Average | general | 56.7% | LiveBench Average Score old version |  |
| Instruction Following Evaluation | instruction-following | 92.1% | Instruction Following Evaluation |  |
| LiveBench Data Analysis | general | 48.3% | LiveBench Data Analysis Score | https://livebench.ai/ |
| LiveBench Language | general | 39.2% | LiveBench Language Score | https://livebench.ai/ |
| Artificial Analysis Quality Index | general | 74 | Artificial Analysis Quality Index (third-party composite, deprecated) |  |
| RankedAGI Math | math | 46.0% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 39.9% | Overall RankedAGI score |  |