Quality
Highest overall scores
- 1
xAI: Grok 4.553.8 - 2xAI: Grok Build 0.139.8
- 3xAI: Grok Latest0
xAI model benchmarks.
Compare xAI models across answer quality, coding, complex tasks, speed, price, and context size.
Response speed
Fastest output in this sample
- 1xAI: Grok Build 0.1115 t/s
- 2xAI: Grok 4.554 t/s
- 3xAI: Grok Latest0 t/s
Input price
Lowest price per 1M tokens
- 1xAI: Grok Build 0.1$1.00/1M
- 2xAI: Grok 4.5$2.00/1M
- 3xAI: Grok Latest$2.00/1M
Highlights
Quality
Overall capability · Higher is better
Speed
Output tokens per second · Higher is better
Input price
USD per 1M tokens · Lower is better
Quality benchmarks
Compare xAI model capability.
Performance and price
Compare practical tradeoffs.
All xAI models.
3 models in the current catalogue.
| Model | Links | |||||
|---|---|---|---|---|---|---|
| xAI: Grok 4.5 | 53.8 | 54 t/s | 500K | $2.00 | $6.00 | |
| xAI: Grok Build 0.1 | 39.8 | 115 t/s | 256K | $1.00 | $2.00 | |
| xAI: Grok Latest | 0 | — | 500K | $2.00 | $6.00 |
xAI technical benchmarks.
Compare named evaluation records available for xAI models. Missing results remain blank rather than being estimated.
| Benchmark | Area | Grok 4.5 | Grok Build 0.1 | Grok Latest | Source |
|---|---|---|---|---|---|
| Intelligence Index | overallQuality | 53.8 | 39.8 | — | OpenRouter / artificial-analysis |
| Coding Index | coding | 72.4 | 51.5 | — | OpenRouter / artificial-analysis |
| Agentic Index | agentic | 45.7 | 28 | — | OpenRouter / artificial-analysis |
| GPQA | GPQA Diamond | 93.1 | 89.5 | — | OpenRouter model benchmarks |
| Humanity's Last Exam | HLE | 40.3 | 36 | — | OpenRouter model benchmarks |
| AA-LCR | AA-LCR | 67.7 | 64.7 | — | OpenRouter model benchmarks |
| GDPval-AA | GDPval-AA | 51.4 | 35.7 | — | OpenRouter model benchmarks |
| CritPt | CritPt | 15.4 | 9.1 | — | OpenRouter model benchmarks |
| SciCode | SciCode | 54.1 | 50.2 | — | OpenRouter model benchmarks |
| AA-Omniscience Accuracy | AA-Omniscience Accuracy | 52 | 51.4 | — | OpenRouter model benchmarks |
| AA-Omniscience Non-Hallucination Rate | AA-Omniscience Non-Hallucination Rate | 46.5 | 9.5 | — | OpenRouter model benchmarks |