Model dossier模型档案
Grok 4.5
Grok 4.5 is a high-tier model from xAI (Intelligence Score 77.9/100). Strongest in knowledge (96.3) and coding (88.6); best fit: knowledge QA and long-form writing. It trails in agent (43.3). Premium-priced: $2.00 in / $6.00 out per 1M tokens.
Grok 4.5 是 xAI 的高水平梯队模型(智能评分 77.9/100)。 最强项是知识(96.3分),其次是代码(88.6分);适合知识问答与长文写作。 短板是智能体(43.3分)。 定价属高端定价:每百万 token 输入 $2.00 / 输出 $6.00。
77.9
Intelligence Score · benchmark coverage 智能评分 · benchmark 覆盖 100%
◈
Benchmark scoresBenchmark 成绩
11 entries 条| Benchmark基准 | Score成绩 | Index指数 | Dated日期 | Measured by测评方 | |
|---|---|---|---|---|---|
| Humanity's Last ExamReasoning | 40.3 | 70 | — | 3rd-party第三方 | source ↗×5⚠±4.1 |
| GPQA DiamondReasoning | 92.9 | 97 | 2026-07-01 | 3rd-party第三方 | source ↗×6⚠±6.5 |
| MMLU-ProKnowledge | 89.2 | 96 | — | 3rd-party第三方 | source ↗×2 |
| SWE-bench VerifiedCoding | 86.6 | 90 | 2026-08 | 3rd-party第三方 | source ↗×7⚠±21.9 |
| AIME 2025Math | 91.7⚡high | 92 | — | 3rd-party第三方 | source ↗×4 |
| LiveCodeBenchCoding | 79 | 84 | 2026-07 | 3rd-party第三方 | source ↗×3⚠±8.35 |
| FrontierMathMath | 19.655 | 21 | — | 3rd-party第三方 | source ↗×2⚠±33.34 |
| Terminal-BenchCoding | 83.3 | 91 | — | official官方榜 | source ↗×4 |
| τ²-benchAgent | 47.94⚡high | 49 | 2026-08-04 | official官方榜 | source ↗ |
| LMArena (Chatbot Arena)Preference | 1469 | 62 | 2026-08-12 | official官方榜 | source ↗×5⚠±50 |
| WebArenaAgent | 35 | 38 | — | 3rd-party第三方 | source ↗×2 |
≡
Specs & pricing规格与价格
| Provider厂商 | xAI |
|---|---|
| Released发布 | 2026-07-08 |
| Context window上下文 | 500K tokens |
| Max output最大输出 | 450K tokens |
| Modality模态 | text+image+file->text |
| Reasoning tiers推理档位 | ⚡low ⚡medium ⚡high default默认 high · low / medium / high (cannot disable) |
| API input price输入价格 | $2.00 / 1M tokens |
| API output price输出价格 | $6.00 / 1M tokens |
⛓
Tools using this model使用它的工具
⚔
Head-to-head正面对比
How we score →评分方法 → · Catalog & pricing via the official OpenRouter API.模型目录与价格来自 OpenRouter 官方 API。