Grok 4.3 API Review (2026): Older Model, Superseded by Grok 4.7
Best for: Readers weighing Grok 4.3's 1M-token context and lower per-token price against the newer Grok 4.7.
Decision summary
Who it’s for, what it costs, and the catch — answered up top.
Bottom line
Older xAI model: 1,000,000-token context, $1.25 input / $2.50 output per 1M tokens. xAI points to Grok 4.7 for code and chat.
Research-based — not hands-on tested. Our scores are editorial judgements compiled from vendor documentation, published pricing and independent user reports. How we review.
Grok 4.3 at a Glance
Grok 4.3 (grok-4.3) is an xAI model that is no longer the current one. xAI’s model guide points developers to Grok 4.7: “For everything else, including code, use Grok 4.7.” xAI’s docs still list grok-4.3 with prices and an alias, grok-4.3-latest, and its model page carries no deprecation notice as of 24 October 2026. This page records what xAI publishes for Grok 4.3 so you can compare it with the current model. Every figure below was read from the xAI page linked beside it on 24 October 2026.
Grok 4.3 and Grok 4.7 Side by Side
Standard rates per 1M tokens, as listed for the us-east-1 region. xAI charges different rates for requests that exceed the 200K context window and lists other regions separately, so check the model pages before budgeting a large-context workload.
| Model | Context window | Input | Cached input | Output |
|---|---|---|---|---|
| Grok 4.3 | 1,000,000 tokens | $1.25 | $0.20 | $2.50 |
| Grok 4.7 | 500,000 tokens | $2.00 | $0.50 | $6.00 |
In the table above, Grok 4.3 has twice the context window and a lower price per token. xAI’s own guidance is still to use Grok 4.7 for code and chat. For prices across providers, see the AI model comparison.
Specifications Listed by xAI
xAI’s Grok 4.3 model page describes the model as a “Fast, reliable model with strong tool calling and instruction following capabilities.”
| Property | xAI’s listing |
|---|---|
| Model name | grok-4.3 (alias grok-4.3-latest) |
| Input | Text and image |
| Output | Text |
| Context window | 1,000,000 tokens |
| Capabilities | Function calling, structured outputs, configurable reasoning |
| Reasoning effort | none, low, medium, high, xhigh (default low) |
| Batch API | Supported |
| Region | us-east-1 |
Real-Time Data Through X Search
xAI’s model guide says Grok has no knowledge of data beyond its training data and tells developers to “enable server-side search tools (Web Search / X Search)” for real-time data. The API pricing page describes X Search as a tool that searches “X posts, user profiles, and threads” and bills it separately from tokens, at $5 / 1k posts and $10 / 1k profiles.
Sources
All pages were read on 24 October 2026.
- Grok 4.3 model page: description, context window, prices, reasoning effort, region
- Grok 4.7 model page: context window and prices
- Models overview: Grok 4.7 recommendation, search-tool guidance
- API pricing: X Search tool pricing
Who it’s for
Ideal for: Readers weighing Grok 4.3's 1M-token context and lower per-token price against the newer Grok 4.7.