Grok-2 Vision: Complete Benchmark Performance Guide

xai 2-vision 2024-08-13 Proprietary

In-depth analysis of Grok-2 Vision performance on MMLU, HumanEval, GSM8K, MATH, BBH benchmarks, including pricing and use cases.

Key Specifications

Vendorxai
Version2-vision
Release Date2024-08-13
Context Window32768 tokens
Input Modalitiestext, image
Output Modalitiestext
LicenseProprietary
Documentationhttps://docs.x.ai/

Benchmark Performance

BenchmarkScoreUnitEvaluated AtNotesSource
MMLU83.4%2024-08-135-shotview
HUMANEVAL84.2pass@12024-08-13—view
GSM8K83.4%2024-08-130-shot CoTview
MATH41.3%2024-08-130-shot CoTview
BBH75.4%2024-08-133-shot CoTview
GPQA43.8%2024-08-130-shotview
IFEVAL76.9%2024-08-13prompt_strictview
ARC93.3%2024-08-13challengeview
MUSR57.7%2024-08-130-shotview
WINOGRANDE87.5%2024-08-130-shotview

Pricing

TierPriceCurrency
Input$2 / MtokUSD
Output$10 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://x.ai/api · as of 2024-08-13

Compliance

  • Data Residency: US
  • SOC2: ✗
  • HIPAA: ✗
  • GDPR: ✗
  • ISO 27001: ✗

Grok-2 Vision

Panoramica del modello

xAI Grok-2 Vision 多模态模型, 32K 上下文, 支持图像理解, 适合视觉问答与多模态推理。

Specifiche principali

FornitoreVersioneData di rilascioFinestra di contestoModalità di inputModalità di outputLicenza
Xai2-vision2024-08-1332Ktext, imagetextProprietary

Prestazioni benchmark

BenchmarkPunteggioUnitàNote
MMLU (Massive Multitask Language Understanding)83.4%5-shot
HumanEval84.2pass@1—
GSM8K (Grade School Math 8K)83.4%0-shot CoT
MATH41.3%0-shot CoT
BBH (BIG-Bench Hard)75.4%3-shot CoT
GPQA43.8%0-shot
IFEval76.9%prompt_strict
ARC93.3%challenge
MUSR57.7%0-shot
WinoGrande87.5%0-shot

Prezzi

InputOutputLettura cacheScrittura cache
————

per milione di token

Punti di forza

  • MMLU score 83.4, strong knowledge reasoning.
  • HumanEval 84.2, excellent code generation.
  • 支持文本、图像、音频多模态输入。

Punti deboli

  • 闭源专有模型，不支持自托管。

Casi d’uso

  • 代码生成与调试
  • 视觉与图像理解

Riferimenti