# Grok 4.3

> xAI's frontier model. 1M context, strong reasoning + tool use.

Human version: https://kymaapi.com/models/grok-4.3
Live JSON: `GET https://kymaapi.com/v1/models` (no auth required)

## Facts

- **Model ID**: `grok-4.3` — pass this as `model` in the request body
- **Creator**: xAI
- **Released**: 2026-04-30
- **Context window**: 1M tokens
- **Max output**: 33K tokens per response — a hard ceiling, not a default
- **Price**: $1.688 in / $3.375 out per 1M, cached input 10%
- **Capabilities**: tools, vision, reasoning, caching

## Call it

```bash
curl https://kymaapi.com/v1/chat/completions \
  -H "Authorization: Bearer $KYMA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "grok-4.3", "messages": [{"role": "user", "content": "Hello"}]}'
```

## Positioning

Grok 4.3 is xAI’s frontier language model, designed for developers building reasoning-heavy applications, agentic workflows, and long-context systems. It is a balanced choice for general-purpose tasks where tool integration and structured outputs are required.

## About

Grok 4.3 is a text-only language model from xAI that supports a 1,000,000-token context window and a maximum output of 32,768 tokens. It is optimized for reasoning, tool use, and structured output generation, making it suitable for chat interfaces, code generation, and autonomous agent pipelines.

On Kyma, the model runs through an OpenAI-compatible endpoint with automatic request failover and exact cost reporting in the usage.cost field. Prompt caching is supported, reducing repeated prefix costs to 10% of the standard input rate. Live telemetry shows a 100% success rate across served requests with a median throughput of 47 tokens per second.

The model does not support vision or multimodal inputs, and its output length is capped at 32,768 tokens. It operates at a medium speed tier, which may introduce latency in high-throughput streaming applications compared to lighter models.

## Use cases

- **Agentic Workflow Orchestration** — Leverages built-in tool calling and reasoning to coordinate multi-step autonomous tasks.
- **Long Context Document Analysis** — Processes up to one million tokens to summarize or query extensive text corpora.
- **Code Generation and Review** — Generates and refines code snippets while maintaining structured output formats.
- **General Purpose Chat Interfaces** — Handles conversational turn-taking with reliable reasoning and balanced latency.

## Not ideal for

It is not suitable for applications requiring image or video understanding, or for workloads that demand high-throughput streaming speeds.

## Pick something else when

- You need faster token throughput → `qwen3.7-flash`
- You need to process images or video → `gemini-3.6-flash`
- You need lower-cost batch processing → `deepseek-v4-flash`

## See also

- All models: https://kymaapi.com/models.md
- Pricing: https://kymaapi.com/pricing.md
- Other models by xAI: https://kymaapi.com/models?q=xAI