xAI
As of 2026-10-18, xAI has 8 models on the meo leaderboard, 2 first-party-tested. Its highest-ranked model is SpaceXAI: Grok 4.7 (#18, 72.4%), while SpaceXAI: Grok 4.3 leads on Effective Value (π).
Official: x.ai Β· huggingface.co/xai-org
| meo rank | Model | Accuracy | π index | $ / correct | Context | Weights |
|---|---|---|---|---|---|---|
| #18 | SpaceXAI: Grok 4.7 | 72.4% | 0.17 | $0.0579 | 500K | β |
| #29 | SpaceXAI: Grok 4.3 | 66.5% | 1.7 | $0.0127 | 1M | β |
| β | SpaceXAI: Grok 4.20 | β | β | β | 2M | β |
| β | SpaceXAI: Grok 4.20 Multi-Agent | β | β | β | 2M | β |
| β | SpaceXAI: Grok 4.3 (batch) | β | β | β | 1M | β |
| β | SpaceXAI: Grok 4.5 | β | β | β | 500K | β |
| β | SpaceXAI: Grok 4.6 | β | β | β | 500K | β |
| β | SpaceXAI: Grok Build 0.1 | β | β | β | 256K | β |
Frequently asked questions
How many AI models does xAI have?
xAI has 8 models on the meo leaderboard, of which 2 are first-party-tested. Its best-ranked model is SpaceXAI: Grok 4.7 at #18 (72.4% accuracy).
What is xAI's best AI model?
On the un-leaked meo benchmark, xAI's top model is SpaceXAI: Grok 4.7, ranked #18 overall with 72.4% accuracy, while SpaceXAI: Grok 4.3 leads the lab on Effective Value (π).
Does xAI release open-weights models?
None of xAI's listed models are flagged open-weights; they are served via API.
Related: all AI labs Β· full leaderboard Β· rankings Β· methodology.