Comprehensive side-by-side LLM comparison
Grok 4 supports multimodal inputs. Both models have their strengths depending on your specific coding needs.
xAI
Grok 4, released by xAI on July 10, 2025, is a large language model featuring first-principles reasoning and comprehensive multimodal support. It features a 260K token context window and demonstrated strong performance on advanced reasoning and coding benchmarks. Grok 4 targets complex multi-step reasoning tasks, scientific analysis, and agentic workflows via the xAI API.
Alibaba / Qwen
Qwen2-72B-Instruct is a 72-billion-parameter language model released by Alibaba's Qwen team in June 2024, serving as the flagship of the Qwen2 generation and representing a major step in open-weight multilingual modeling. Trained on data spanning 30+ languages with strong coverage of code and structured reasoning, the model was among the first openly released 70B-class models to demonstrate competitive performance across diverse benchmarks. It established the foundation architecture and training methodology that the Qwen2.5 series would later extend.
1 year newer
Qwen2 72B Instruct
Alibaba / Qwen
2024-06-07

Grok 4
xAI
2025-07-10
Context window and performance specifications
Available providers and their performance metrics
Grok 4
xAI
Qwen2 72B Instruct
Grok 4
Qwen2 72B Instruct
Grok 4
Qwen2 72B Instruct