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

Benchmark Score Unit Evaluated At Notes Source
MMLU 83.4 % 2024-08-13 5-shot view
HUMANEVAL 84.2 pass@1 2024-08-13 — view
GSM8K 83.4 % 2024-08-13 0-shot CoT view
MATH 41.3 % 2024-08-13 0-shot CoT view
BBH 75.4 % 2024-08-13 3-shot CoT view
GPQA 43.8 % 2024-08-13 0-shot view
IFEVAL 76.9 % 2024-08-13 prompt_strict view
ARC 93.3 % 2024-08-13 challenge view
MUSR 57.7 % 2024-08-13 0-shot view
WINOGRANDE 87.5 % 2024-08-13 0-shot view

Pricing

Tier Price Currency
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

Model Overview

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。

Core Specifications

Vendor Version Release Date Context Window Input Modalities Output Modalities License
Xai 2-vision 2024-08-13 32K text, image text Proprietary

Benchmark Performance

Benchmark Score Unit Notes
MMLU (Massive Multitask Language Understanding) 83.4 % 5-shot
HumanEval 84.2 pass@1 —
GSM8K (Grade School Math 8K) 83.4 % 0-shot CoT
MATH 41.3 % 0-shot CoT
BBH (BIG-Bench Hard) 75.4 % 3-shot CoT
GPQA 43.8 % 0-shot
IFEval 76.9 % prompt_strict
ARC 93.3 % challenge
MUSR 57.7 % 0-shot
WinoGrande 87.5 % 0-shot

Pricing

Input Output Cache Read Cache Write
— — — —

per million tokens

Strengths

  • MMLU score 83.4, strong knowledge reasoning.
  • HumanEval 84.2, excellent code generation.
  • Input Modalities: text, image, audio.

Weaknesses

  • Proprietary, not self-hostable.

Use Cases

  • Code generation and debugging
  • Vision and image understanding

References