Update xAI LLM Proprietary

xAI Grok-1.5

Grok's reasoning upgrade — first xAI model with strong STEM benchmarks

Released 2024-03-28 · ~314B · 128K tokens context

Overview

Enhanced version of Grok with improved performance across benchmarks

Specifications

DeveloperxAI
Release date2024-03-28
Model typeLLM
Parameters~314B
ArchitectureTransformer
Context window128K tokens
LicenseProprietary
Input modalitiestext
Output modalitiestext

Benchmarks

Benchmark Grok-1.5 Grok-1Δ
MMLU 73.0% 73.0% 0.0
HumanEval 63.2% 63.2% 0.0
MATH 50.6%

Availability

Platform Exclusive xAI API

What's new vs Grok-1

Direct comparison available in the model comparison tool →

Official references