Grok 4 Models by xAI
Details
Capabilities
About
xAI's Grok 4 model family spans the original Grok 4 API lineup, Grok 4.3, Grok 4.20 variants, and Grok 4.5 — the July 2026 flagship for coding, agentic tasks, and knowledge work.
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
Use when the workload needs code, 500k context, and reasoning.
Use when the workload needs code, 256k context, and reasoning.
Use when the workload needs 1m context, reasoning, and tool use.
Use when the workload needs agentic, 1m context, and reasoning.
Use when the workload needs reasoning, 1m context, and tool use.
Use when the workload needs 1m context, tool use, and function calling.
Use when the workload needs 1m context, reasoning, and tool use.
Use when the workload needs 131k context, reasoning, and tool use.
Use when the workload needs 256k context and multimodal inputs.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Grok 4.5 | Use when the workload needs code, 500k context, and reasoning. | 2026-07 | code500k contextreasoning | Current |
| Grok Build 0.1 | Use when the workload needs code, 256k context, and reasoning. | 2026-05 | code256k contextreasoning | Current |
| Grok 4.3 | Use when the workload needs 1m context, reasoning, and tool use. | 2026-05 | 1m contextreasoningtool use | Current |
| Grok 4.20 Multi-Agent | Use when the workload needs agentic, 1m context, and reasoning. | 2026-03 | agentic1m contextreasoning | Current |
| Grok 4.20 Reasoning | Use when the workload needs reasoning, 1m context, and tool use. | 2026-03 | reasoning1m contexttool use | Current |
| Grok 4.20 Non-Reasoning | Use when the workload needs 1m context, tool use, and function calling. | 2026-03 | 1m contexttool usefunction calling | Current |
| Grok 4.20 | Use when the workload needs 1m context, reasoning, and tool use. | 2026-02 | 1m contextreasoningtool use | Current |
| Grok 4.1 | Use when the workload needs 131k context, reasoning, and tool use. | 2025-11 | 131k contextreasoningtool use | Current |
| Grok 4 Heavy | Use when the workload needs 256k context and multimodal inputs. | 2025-07 | 256k contextmultimodal inputs | Current |
Release Timeline
8 release groupsReplaced By
Keep for legacy integrations; evaluate Grok 4.3 before new work.
Keep for legacy integrations; evaluate Grok 4.3 before new work.
Keep for legacy integrations; evaluate Grok 4.3 before new work.
Keep for legacy integrations; evaluate Grok Build 0.1 before new work.
Specifications(14 models)
| Model | Released | Context | Parameters | Vision | Multimodal | Reasoning | Fn Calling | Tool Use | Structured Outputs | Code Exec |
|---|---|---|---|---|---|---|---|---|---|---|
| Grok 4.5 | 2026-07 | 500k | — | Yes | Yes | Yes | Yes | Yes | No | Yes |
| Grok Build 0.1 | 2026-05 | 256k | — | Yes | Yes | Yes | Yes | Yes | Yes | No |
| Grok 4.3 | 2026-05 | 1m | ~0.5T | Yes | Yes | Yes | Yes | Yes | Yes | No |
| Grok 4.20 Multi-Agent | 2026-03 | 1m | — | Yes | Yes | Yes | Yes | Yes | Yes | No |
| Grok 4.20 Reasoning | 2026-03 | 1m | — | Yes | Yes | Yes | Yes | Yes | Yes | No |
| Grok 4.20 Non-Reasoning | 2026-03 | 1m | — | Yes | Yes | No | Yes | Yes | Yes | No |
| Grok 4.20 | 2026-02 | 1m | — | Yes | Yes | Yes | Yes | Yes | Yes | No |
| Grok 4.1 | 2025-11 | 131k | — | Yes | Yes | Yes | Yes | Yes | No | No |
| Grok 4 Heavy | 2025-07 | 256k | — | Yes | Yes | No | No | No | No | No |
Available From(7 providers)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| Grok Build 0.1 | xAI Console | $1 | $2 | Serverless |
| Grok Build 0.1 | Vercel AI Gateway | $1 | $2 | Serverless |
| Grok Build 0.1 | OpenRouter | $1 | $2 | Serverless |
| Grok 4.20 | xAI Console | $1.25 | $2.5 | Serverless |
| Grok 4.3 | xAI Console | $1.25 | $2.5 | Serverless |
| Grok 4.20 Multi-Agent | xAI Console | $1.25 | $2.5 | Serverless |
| Grok 4.3 | OpenRouter | $1.25 | $2.5 | Serverless |
| Grok 4.20 | OpenRouter | $1.25 | $2.5 | Serverless |
| Grok 4.20 Reasoning | xAI Console | $1.25 | $2.5 | Serverless |
| Grok 4.20 Non-Reasoning | xAI Console | $1.25 | $2.5 | Serverless |
| Grok 4.20 Multi-Agent | Vercel AI Gateway | $1.25 | $2.5 | Serverless |
| Grok 4.20 Non-Reasoning | Vercel AI Gateway | $1.25 | $2.5 | Serverless |
| Grok 4.20 Reasoning | Vercel AI Gateway | $1.25 | $2.5 | Serverless |
| Grok 4.3 | Vercel AI Gateway | $1.25 | $2.5 | Serverless |
| Grok 4.3 | AWS Bedrock | $1.25 | $2.5 | Serverless |
| Grok 4.20 Multi-Agent | OpenRouter | $2 | $6 | Serverless |
| Grok 4.5 | xAI Console | $2 | $6 | Serverless |
| Grok 4.5 | OpenRouter | $2 | $6 | Serverless |
| Grok 4.5 | Vercel AI Gateway | $2 | $6 | Serverless |
Popular comparisons in this family
Comparisons
- GPT-4o (08-06) vs Grok 4
- o3 vs Grok 4
- Grok 4 Fast Reasoning vs o3
- Grok 4.3 vs GPT-5.5
- Claude Sonnet 4.6 vs Grok 4
- Claude Opus 4.6 vs Grok 4
- Grok 4 Fast Reasoning vs Claude Opus 4.6
- Claude Fable 5 vs Grok 4
Frequently Asked Questions
- What is Grok 4 used for?
- Grok 4 is used for code, agentic, and reasoning. The family description and listed model capabilities point to those workloads as the best fit.
- How does Grok 4 compare to Grok 1?
- Grok 4 by xAI is strongest where you need code, while Grok 1 by xAI is the closest related family to check for reasoning. Grok 4 has 14 listed variants and reaches up to 2m context, so compare the specs and pricing tables before choosing a production model.
- Which Grok 4 model should I use?
- For the lowest listed input price, start with Grok 4.1 Fast through OpenRouter at $0.2/1M input tokens. For the most capable/latest local choice, evaluate Grok 4.3 with 1m context and reasoning, tool use, function calling, structured outputs, and multimodal inputs.
Models(14)
Grok 4.5
Grok Build 0.1
Grok 4.3
Grok 4.20 Multi-Agent
Grok 4.20 Reasoning
Grok 4.20 Non-Reasoning
Grok 4.20
Grok 4.1
Grok 4 Heavy





