xAI
xAI
MultimodalLIVE PRICING

SpaceXAI: Grok 4.20 Multi-Agent

text+image+file->text

The Grok 4 generation model that preceded the 4.3 flagship, with a 256K-token context window. Grok 4.3 supersedes it on both capability and price, so this entry is mainly relevant to workloads already pinned to Grok 4 behaviour.

SpaceXAI: Grok 4.20 Multi-Agent is a multimodal AI model from xAI. It costs $1.250 per million input tokens and $2.500 per million output tokens (blended $1.625/M), with a 2,000,000-token context window.

Profile inherited from upstream Grok 4 ↗ — this is a hosted variant of the same open-weights model.

INPUT
$1.250/M
per million input tokens
OUTPUT
$2.500/M
per million output tokens
BLENDED 70/30
$1.625/M
unchanged since 3 May
CONTEXT
2,000,000
tokens
What it is good at
  • 256K-token context
  • Established behaviour for existing pipelines
Typical use cases
  • Pipelines pinned to Grok 4
  • Migration baselines

Benchmarks

No published benchmark scores tracked for this model yet. Frontier and reasoning models from major providers have scores; smaller models and inference-host variants typically inherit the underlying open-weights score.

More from xAI

See all 8 →

Frequently asked questions

How much does SpaceXAI: Grok 4.20 Multi-Agent cost?

SpaceXAI: Grok 4.20 Multi-Agent costs $1.250 per million input tokens and $2.500 per million output tokens, for a blended reference rate of $1.625 per million tokens.

What is SpaceXAI: Grok 4.20 Multi-Agent's context window?

SpaceXAI: Grok 4.20 Multi-Agent supports up to 2,000,000 tokens of context in a single request.

What is SpaceXAI: Grok 4.20 Multi-Agent best for?

SpaceXAI: Grok 4.20 Multi-Agent is well suited to 256K-token context, Established behaviour for existing pipelines and Pipelines pinned to Grok 4.

Who makes SpaceXAI: Grok 4.20 Multi-Agent?

SpaceXAI: Grok 4.20 Multi-Agent is developed and served by xAI.

Terms used on this page