Google vs 
SpaceXAI
Verdict
- SpaceXAI's flagship leads: Grok 4.6 at 66.0 vs Gemini 3.8 Flash at 64.8.
- Google's fastest ranked model (Gemini 3.5 Flash Lite, 392 tok/s) beats SpaceXAI's (Grok 4.20 Multi-Agent, 304 tok/s).
- Google publishes more open-weights models (15 vs 1).
- Google ships more often: one model every 13 days against every 42.
Best model per price band
Each provider’s highest-index model below the price ceiling. Purple marks the winner of the band.
| Price band | SpaceXAI | |
|---|---|---|
| Under $0.50/M | Gemma 4 26B A4B 57.1$0.198/M | – |
| Under $2/M | Gemini 3.8 Flash medium 64.8$1.50/M | Grok 4.20 thinking 59.9$1.56/M |
| Under $10/M | Gemini 3.8 Flash medium 64.8$1.50/M | Grok 4.6 medium 66.0$3.00/M |
| Any price | Gemini 3.8 Flash medium 64.8$1.50/M | Grok 4.6 medium 66.0$3.00/M |
Line-up
| Metric | SpaceXAI | |
|---|---|---|
| Flagship index | 64.8 | 66.0 |
| Ranked models | 40 | 16 |
| Open-weights models | 15 | 1 |
| Cheapest ranked, $/M | $0.040 | $1.25 |
| Fastest ranked, tok/s | 392 | 304 |
| Release cadence, days | 13 | 42 |
| Last release | 2 Sept 2026 | 21 Sept 2026 |
| Next expected | 9 Sept 2026 – 20 Sept 2026 | 16 Oct 2026 – 18 Nov 2026 |