Grok 4.3
Grok 4.3 is xAI's multimodal model with a 1M-token context window, agentic tool use, vision support, and structured outputs for enterprise workflows.
Model Overview
Capabilities, design details, and architectural traits
Grok 4.3 - Frontier Multimodal Model for Long-Context
Grok 4.3 (grok-4.3) is xAI's multimodal language model designed for long-context reasoning, enterprise AI agents, and production-scale applications. It combines a 1-million-token context window with native tool use, structured outputs, image understanding, and function calling, making it suitable for retrieval-augmented generation, software development, and complex analytical workloads.
Documented Characteristics
- 1M-token context window for long documents and extended conversations.
- Native support for text and image inputs with text generation.
- Built-in function calling, structured JSON outputs, and agentic tool use.
- Designed for complex reasoning, coding, planning, and enterprise AI workflows.
- Supports web search and external tool integration through compatible platforms.
- Proprietary commercial model available through the xAI API and partner platforms.
Benchmark Performance
Independent evaluations · Artificial Analysis
Accuracy & Capability Details
Compare Models Side-by-Side
Evaluate specifications, pricing, and independent benchmark indices
| Model Details | |||
|---|---|---|---|
| General Info | |||
| Provider | xAI | Anthropic | Anthropic |
| Release Date | April 30, 2026 | September 22, 2026 | September 28, 2026 |
| Knowledge Cutoff | - | - | Jun 2026 |
| Context & Limits | |||
| Context Window | 1M | 1M | 1M |
| Pricing (per 1M tokens) | |||
| Input Pricing | $1.25 Best Input Pricing | $4 | $2 |
| Output Pricing | $2.50 Best Output Pricing | $20 | $10 |
| Modalities | |||
| Inputs | textimagefile | textimagefile | textimagefile |
| Outputs | text | text | text |
| Benchmarks (0-100) | |||
| Intelligence Index | 24.9 | 57.6 Best Intelligence Index | 56.0 |
| Coding Index | 42.2 | - | - |
| Agentic Index | 15.5 | - | - |
Humanity's Last Exam
Extremely difficult logical reasoning and knowledge.
Long Context Reasoning
Logical reasoning over long context windows.
SciCode Benchmark
Scientific coding and mathematical modeling.
Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.
Explore more from xAI
Other models by xAI
Top AI Models
Leading alternatives by intelligence score
