Kyma APIKyma API
xAI

xAI

Grok 4.3

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.

Modalities

Text+Image → Text

Below release

Not established

Weights

Not published

Where it sits in the catalogue Kyma measures

Every number against every text model Kyma prices per token — a stated rule, not a chosen line-up.

Price · input#52/71

$1.92

output $3.838 · cached $0.27

median $1.013 · best $0.0494

Price · output#40/70

$3.838

per 1M tokens generated

median $3.531 · best $0.218

Availability · 30d#50/70

93.18%

44 observations since 2026-05-24

median 96.5% · best 100.0%

Throughput#13/67

73 tok/s

probe median, one fixed prompt

median 36.0 tok/s · best 363.2 tok/s

Response time#34/67

2.67 s

probe median, to a complete answer

median 2.67 s · best 0.83 s

Context

1M

max output 32,768

published by xAI, not measured here

Tick above each rail is this model, below it the other 70. Dashed rule is the field median, solid is its best. Better is left; the axis stops at the 90th percentile, so a few models sit past its right edge.

Usage

How much this model is actually called here.

Rank

#39

of 100 active models

Tokens served

4.5M

all-time

Platform share

0.1%

of all tokens

Pricing

Pay per token. Cached input bills at this model’s own cached rate, listed below.

$1.92 /1M input$3.838 /1M output
Hobby10 req/day · 2K in / 500 out
~$1.73/mo
Production1,000 req/day · 2K in / 500 out
~$173/mo
Scale20,000 req/day · 2K in / 500 out
~$3,455/mo
+ Estimate your workload
1,000
2,000
500
30%

Estimated monthly cost

$143

$4.77 / day on Grok 4.3

Same workload on:

Grok 4.7$227+59%
Grok 4.6$290+102%

Estimates use list pricing with cached input at this model's own cached rate. Actual bills depend on real token counts, and every response includes its exact cost.

When to use Grok 4.3

Updated 2026-07-31

Where this model earns its cost — and where it doesn't.

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 this model's cached input rate.

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.

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.

How it compares

Against the peers people actually weigh it against.

SpecGrok 4.3Grok 4.7Grok 4.6
Input /1M$1.92$2.16$2.758
Output /1M$3.838$6.48$8.274
Context1M500K500K
ToolsYesYesYes
ReasoningYesYesYes
Throughput73 tok/s59 tok/s

Quick start

Up and running in under two minutes.

  1. 1

    Create an API key

    Sign up and grab a key from the dashboard. Grok 4.3 needs a top-up — the signup credit covers the free tier.

    Get API key →
  2. 2

    Make your first request

    Drop in your key and send a chat completion — fully OpenAI-compatible.

    curl https://kymaapi.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok-4.3",
    "messages": [
      {"role": "user", "content": "Explain prompt caching in one paragraph."}
    ]
  }'
  3. 3

    Stream responses

    Add "stream": true to receive tokens as they arrive.

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

FAQ

Common questions about this model.

What is the context window of Grok 4.3?

Grok 4.3 has a 1M-token context window — roughly 1471 pages of text in a single request.

How much does the Grok 4.3 API cost?

$1.92 per 1M input tokens and $3.838 per 1M output tokens, with cached input at $0.27/1M — 86% off the input rate on repeated prompt prefixes. No subscription; you pay only for what you use.

Does Grok 4.3 support function calling?

Yes — Grok 4.3 supports tool/function calling and structured outputs (JSON mode), so it works with agent frameworks out of the box.

Are the weights for Grok 4.3 publicly available?

No. xAI does not publish Grok 4.3's weights — it is reachable through an API rather than as a file you can download and run (https://docs.x.ai/developers/models/grok-4.3, read 2026-08-14). That is xAI's decision about their own product, not a limitation of Kyma.

Is Grok 4.3 ever served below the precision its creator released it at?

Kyma cannot say, and for this model nobody can. xAI does not publish Grok 4.3's weights, so there is no released checkpoint whose precision a route could be measured against. We publish "not established" rather than reading a low format as a downgrade: on several models in this catalogue a low format is exactly how the creator ships it.

How do I use Grok 4.3?

Kyma is OpenAI-compatible: point your SDK's base URL at https://kymaapi.com/v1, use your Kyma API key, and set the model to grok-4.3. Signing up is free, but this model needs a top-up: the $0.50 signup credit covers the free tier only.

Does Grok 4.3 support image inputs?

No, the model accepts text-only inputs and does not support vision or multimodal processing.

How does prompt caching work on Kyma for this model?

Repeated prompt prefixes are billed at this model's cached input rate, automatically applied when caching is enabled.

What happens if the serving path experiences degradation?

Kyma automatically reroutes the request to a healthy serving path, ensuring consistent availability without manual intervention.

Needs a top-up — the signup credit covers the free tier.Create account →

More models by xAI

ModelContextInputOutput
xAIGrok 4.7500K$2.16$6.48
xAIGrok 4.6500K$2.758$8.274
xAIGrok 4.5500K$2.73$8.189
xAIGrok Build256K$1.389$2.777
xAIGrok 4.202M$1.763$3.528
xAIGrok 4.20 Multi-Agent2M$1.767$3.534