UltraGPT
SpaceXAI: Grok 4.20 Multi-Agent
Grok 4.20 Multi-Agent is a variant of SpaceXAI’s Grok 4.20 designed for collaborative, agent-based workflows. Multiple agents operate in parallel to conduct deep research, coordinate tool use, and synthesize information...
Intelligence overview
2M tok
Context window
1.80M tok
Max output
What SpaceXAI: Grok 4.20 Multi-Agent can do.
Reasoning
Extended reasoning
Works through problems step by step before answering, trading latency for accuracy on harder tasks.
Adjustable effort
Choose how much the model deliberates per request — trade speed for depth as the task demands.
Multimodal
Image understanding
Reads and reasons over images alongside text in the same prompt.
File input
Reads uploaded files — documents, spreadsheets, code — directly in context.
Development
Structured output
Returns responses constrained to a JSON schema, ready to parse without cleanup.
Streaming
Streams tokens as they're generated instead of waiting on the full response.
Pay for what you use.
Every model in UltraGPT runs behind one flat subscription — this is the underlying per-token cost the catalog tracks.
$1.25 / 1M tok
$2.5 / 1M tok
Built for serious work.
Build
Generate and refactor code, with the ability to return structured, parseable output.
Think
Works through hard problems with adjustable reasoning effort — xhigh, high, medium, low.
Research
Holds up to 2M tokens of source material in context without losing the thread.
Create
Reads images alongside a prompt to ground creative and design work in what's actually on screen.
Ask SpaceXAI: Grok 4.20 Multi-Agent anything.
A preview of the real UltraGPT composer — the model you see here is the model you get.
Continues in the UltraGPT app — one login, every model.
How SpaceXAI: Grok 4.20 Multi-Agent stacks up.
The closest models in the catalog by provider, capability and score — real entries, pulled from the same directory.
Technical details.
Specification
- Context window
- 2M tokens
- Max output
- 1.80M tokens
- Input modalities
- Text + Image + File
- Output modalities
- Text
- Knowledge cutoff
- Sep 1, 2025
- Release date
- Mar 31, 2026
- Reasoning effort
- xhigh, high, medium, low (always on)
- Streaming
- Supported
- Tool calling
- Not supported
- JSON mode
- Supported
Reasoning
More from xAI, and everything else.
This model is one entry in the full UltraGPT catalog — browse the rest, compare pricing and context windows, or filter by capability.