xAI
SpaceXAI: Grok 4.3
x-ai/grok-4.3
Grok 4.3 is a reasoning model from SpaceXAI. It accepts text and image inputs with text output, and is suited for agentic workflows, instruction-following tasks, and applications requiring high factual accuracy. Reasoning can be configured between none/low/medium/high (default low) effort levels. It supports a 1 million token context window with no output token limit, making it well-suited for long-document analysis, deep research, and multi-step agentic tasks. Pricing is tiered: requests exceeding 200k total tokens are billed at a higher rate.
/ 1M tokens
/ 1M tokens
tokens
Limits
Context
1,000,000
tokens
Max output
900,000
90% of the context window
Model facts
Author
xAI
Tokenizer
Grok
Modality
Text + Image + File -> Text
Created
Apr 30, 2026
Reasoning
Publication status
Modalities
Generation controls
Capability scores
Better than 40% of models compared
Better than 36% of models compared
Better than 61% of models compared
Agent tasks
Category performance
1118 Elo
43.2% win
1055 Elo
33.6% win
1071 Elo
32.5% win
1068 Elo
32.4% win
1072 Elo
31.9% win
1066 Elo
31.8% win
Additional published result
Agentic slides (HTML)
1066 Elo · Rank #9 · 31.7% win · Top 94%
1072 Elo
30.8% win
1007 Elo
28.8% win
1008 Elo
28.4% win
1006 Elo
28.1% win
976 Elo
22.7% win
914 Elo
20.7% win
Creative tasks
Category performance
1180 Elo
45.2% win
1193 Elo
45.0% win
1197 Elo
44.2% win
1189 Elo
43.8% win
1176 Elo
43.4% win
1131 Elo
43.1% win
1126 Elo
40.9% win
1080 Elo
37.2% win