Compare AI models
| Specifications | ||
Context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider. | 32k | 256k |
| Overview | ||
| Company | SpaceXSI | |
| Release date | Feb 8 2024 | Jul 9 2025 |
| Access | Closed | Closed |
| Model details | View model | View model |
Other comparisons
Gemini 1.0 UltravsClaude Haiku 5.5Grok 4vsClaude Haiku 5.5Gemini 1.0 UltravsGPT-6.1 SolGrok 4vsGPT-6.1 SolGemini 1.0 UltravsMuse Spark OpenGrok 4vsMuse Spark OpenGemini 1.0 UltravsDeepSeek-V4.1-Flash-1017Grok 4vsDeepSeek-V4.1-Flash-1017Gemini 1.0 UltravsMistral Large 4Grok 4vsMistral Large 4Gemini 1.0 UltravsKimi K3Grok 4vsKimi K3Frequently asked questions
Gemini 1.0 Ultra and Grok 4 don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Gemini 1.0 Ultra shipped 517 days before Grok 4, so benchmark comparisons should account for the intervening progress.
Context windows are 32k (Gemini 1.0 Ultra) vs 256k (Grok 4).
Direct benchmark comparisons are unavailable — Gemini 1.0 Ultra and Grok 4 don't publish scores on any of the same benchmarks.