SpaceXAI发布Grok 4.7：更大基座模型，价格不变
SpaceXAI Releases Grok 4.7: A Larger Base Model at the Same $2/$6 Price as Grok 4.6
Grok 4.7在保持低价的同时显著提升了编码与安全能力，对Agent开发有直接参考价值，建议开发者关注其性能表现。
SpaceXAI has released Grok 4.7, its new flagship model for coding, agentic tasks, and knowledge work. Grok 4.7 is built on a larger base model and a longer reinforcement learning run. It still ships at the same price and speed as Grok 4.6.
SpaceXAI 发布了 Grok 4.7，这是其用于编码、智能体任务和知识工作的全新旗舰模型。Grok 4.7 基于更大的基础模型和更长的强化学习训练周期构建。它仍保持与 Grok 4.6 相同的价格和速度。
Is it deployable? Yes, as a hosted model. You can call grok-4.7 today through the xAI API, Cursor, Grok Build, OpenRouter, Vercel, and Cloudflare.
可以部署吗？可以，作为托管模型使用。你今天就可以通过 xAI API、Cursor、Grok Build、OpenRouter、Vercel 和 Cloudflare 调用 grok-4.7。
What Changed Under the Hood
底层架构的变化
SpaceXAI lists 4 changes over Grok 4.6:
SpaceXAI 列出了相较于 Grok 4.6 的 4 项变化：
- A new, larger base model: Grok 4.7 does not reuse the Grok 4.6 base.
- A longer RL run on harder tasks: The task mix is weighted toward problems that take many hours to complete.
- Better self-verification and long-context handling: The company says the model checks its own work more carefully.
- Native Grok Bot harness support: It was trained to understand the Grok Bot harness for conversational and knowledge work.
- 新的更大规模基础模型：Grok 4.7 没有复用 Grok 4.6 的基础模型。
- 在更难的任务上进行更长的 RL 训练：任务组合加权偏向于需要数小时才能完成的问题。
- 更好的自我验证和长上下文处理能力：该公司表示，该模型会更仔细地检查自己的工作。
- 原生 Grok Bot harness 支持：它经过训练以理解用于对话和知识工作的 Grok Bot harness。
The developer docs list the API specs:
开发者文档列出了 API 规格：
| Property | Value |
|---|---|
| Model name | grok-4.7 |
| Context window | 500,000 tokens |
| Knowledge cutoff | May 2026 |
| Modalities | Text and image input, text output |
| Reasoning effort | low, medium, high (default), xhigh |
| APIs | Responses API, Chat Completions |
| Tools | Function calling, web search, X search, code execution |
| 属性 | 值 |
|---|---|
| 模型名称 | grok-4.7 |
| 上下文窗口 | 500,000 tokens |
| 知识截止日期 | 2026年5月 |
| 模态 | 文本和图片输入，文本输出 |
| 推理努力程度 | low（低）、medium（中）、high（高，默认）、xhigh（极高） |
| APIs | Responses API、Chat Completions |
| 工具 | 函数调用、网络搜索、X 搜索、代码执行 |
Benchmarks
基准测试
The launch table compares Grok 4.7 at xHigh effort with Grok 4.6 High, GPT-5.6 Sol Max, and Fable 5.1 Max. The Grok 4.7 DeepSWE score was run at high effort. All scores are vendor-reported.
发布表格比较了 xHigh 努力程度下的 Grok 4.7 与 Grok 4.6 High、GPT-5.6 Sol Max 以及 Fable 5.1 Max。Grok 4.7 的 DeepSWE 分数是在 high 努力程度下运行的。所有分数均由供应商报告。
| Benchmark | Grok 4.7 xHigh | Grok 4.6 High | GPT-5.6 Sol Max | Fable 5.1 Max |
|---|---|---|---|---|
| Input price ($/M) | $2 | $2 | $4 | $10 |
| Output price ($/M) | $6 | $6 | $20 | $50 |
| CursorBench 4.0 | 46.3% | 40.4% | 41.7% | 51.8% |
| DeepSWE v1.1 | 71.0%* | 65.2% | 72.7% | 70.0% |
| EEBench | 64.0% | 53.0% | 39.4% | 56.4% |
| AA Briefcase v1.1 | 1,657 | 1,546 | 1,487 | 1,678 |
| Terminal-Bench 4.0 | 38.0% | 20.3% | 37.3% | 57.9% |
| Harvey Legal Agent Benchmark | 19.6% | 15.8% | 2.5% | 6.7% |
| HealthBench Professional | 56.7% | 48.5% | 60.5% | 62.1% |
| 基准测试 | Grok 4.7 xHigh | Grok 4.6 High | GPT-5.6 Sol Max | Fable 5.1 Max |
|---|---|---|---|---|
| 输入价格 ($/M) | $2 | $2 | $4 | $10 |
| 输出价格 ($/M) | $6 | $6 | $20 | $50 |
| CursorBench 4.0 | 46.3% | 40.4% | 41.7% | 51.8% |
| DeepSWE v1.1 | 71.0%* | 65.2% | 72.7% | 70.0% |
| EEBench | 64.0% | 53.0% | 39.4% | 56.4% |
| AA Briefcase v1.1 | 1,657 | 1,546 | 1,487 | 1,678 |
| Terminal-Bench 4.0 | 38.0% | 20.3% | 37.3% | 57.9% |
| Harvey Legal Agent Benchmark | 19.6% | 15.8% | 2.5% | 6.7% |
| HealthBench Professional | 56.7% | 48.5% | 60.5% | 62.1% |
*High effort
*高努力程度
Grok 4.7 improves on Grok 4.6 in every row. The largest jump is on Terminal-Bench 4.0, from 20.3% to 38.0%. EEBench rose 11 points to 64.0%, the top score in the table. On Harvey’s legal agent benchmark, Grok 4.7 scored 19.6% against 6.7% for Fable 5.1 Max.
Grok 4.7 在所有行上均优于 Grok 4.6。最大的提升出现在 Terminal-Bench 4.0，从 20.3% 跃升至 38.0%。EEBench 提升了 11 分至 64.0%，成为表中的最高分。在 Harvey 的法律代理基准测试中，Grok 4.7 得分为 19.6%，而 Fable 5.1 Max 为 6.7%。
Grok 4.7 does not lead across the board. Fable 5.1 Max tops 4 of 7 benchmarks, including a 57.9% Terminal-Bench score. GPT-5.6 Sol Max holds the top DeepSWE v1.1 result at 72.7%.
Grok 4.7 并未在所有方面领先。Fable 5.1 Max 在 7 项基准测试中的 4 项排名第一，包括 57.9% 的 Terminal-Bench 得分。GPT-5.6 Sol Max 在 DeepSWE v1.1 上保持 72.7% 的最高结果。
Price is the other axis. Fable 5.1 Max costs 5x more on input and about 8.3x more on output. GPT-5.6 Sol Max costs 2x more on input and about 3.3x more on output. On a CursorBench 4.0 cost-per-task chart, SpaceXAI places Grok 4.7 at the frontier in price-performance.
价格是另一个维度。Fable 5.1 Max 的输入成本高出 5 倍，输出成本高出约 8.3 倍。GPT-5.6 Sol Max 的输入成本高出 2 倍，输出成本高出约 3.3 倍。在 CursorBench 4.0 的任务成本图表上，SpaceXAI 将 Grok 4.7 置于性价比的前沿位置。
On GDPval, which tests professional knowledge work, Grok 4.7 xhigh scored 1,695 Elo. That is up from 1,605 for Grok 4.6 high. Fable 5.1 max leads at 1,735, and GPT-6 Astra max scored 1,542. SpaceXAI also says Grok 4.7 is better at creating documents and presentations.
在测试专业知识工作的 GDPval 上，Grok 4.7 xhigh 得分 1,695 Elo。这高于 Grok 4.6 high 的 1,605 分。Fable 5.1 max 以 1,735 分领先，GPT-6 Astra max 得分为 1,542。SpaceXAI 还表示，Grok 4.7 在创建文档和演示文稿方面表现更佳。
Safety and Cybersecurity
安全与网络安全
Grok 4.7 ships with an entirely new safeguard stack. SpaceXAI calls it the strongest model it has tested on refusals and jailbreak resistance. It topped LatchBio’s biosafety benchmark at 62.4%.
Grok 4.7 配备了全新的安全防护堆栈。SpaceXAI 称这是其在拒绝响应和抵抗越狱方面测试过的最强模型。它在 LatchBio 的生物安全基准测试中以 62.4% 的成绩位居榜首。
On HackerBench v0.3, SpaceXAI’s own benchmark for risky and malicious cyber tasks, the model let 3.3% of risky dual-use prompts through. The company says it rarely blocks legitimate security work. Select cybersecurity partners now get invite-only access to its red-team capabilities for defense research.
在 SpaceXAI 自身针对高风险和恶意网络任务的 HackerBench v0.3 基准测试中，该模型放行了 3.3% 的高风险双重用途提示词。该公司表示，它很少阻止合法的安全工作。目前，部分网络安全合作伙伴获得了对其红队能力的邀请制访问权限，用于防御研究。
Pricing and Availability
定价与可用性
Grok 4.7 costs $2 per million input tokens and $6 per million output tokens. It is available in Cursor on all plans and is the default model in Grok Build. It is also served through the Grok API, OpenRouter, Vercel, and Cloudflare.
Grok 4.7 的输入 token 价格为每百万 2 美元，输出 token 价格为每百万 6 美元。它在所有计划的 Cursor 版本中均可用，并且是 Grok Build 中的默认模型。它还通过 Grok API、OpenRouter、Vercel 和 Cloudflare 提供服务。
Grok 4.7 Fast is the same model on faster infrastructure, with twice the output speed at twice the price. The docs say it runs only in Cursor and Grok Build, not on the public xAI API. It is also excluded from Grok Build’s free tier.
Grok 4.7 Fast 是在更快基础设施上运行的同一模型，输出速度翻倍，价格也是两倍。文档指出，它仅在 Cursor 和 Grok Build 中运行，不在公共 xAI API 上使用。它也排除在 Grok Build 的免费层级之外。
A US regional endpoint at https://us.api.x.ai/v1 keeps inference in the United States at a 10% premium. SpaceXAI recommends setting a prompt_cache_key for reliable cache hits.
位于 https://us.api.x.ai/v1 的美国区域端点可将推理保留在美国境内，但需支付 10% 的溢价。SpaceXAI 建议设置 prompt_cache_key 以确保可靠的缓存命中。
import os
from xai_sdk import Client
from xai_sdk.chat import user
client = Client(api_key=os.getenv("XAI_API_KEY"))
chat = client.chat.create(model="grok-4.7")
chat.append(user("Explain this repo."))
print(chat.sample().content)Interactive Explainer
交互式解释器
Key Takeaways
关键要点
- Grok 4.7 keeps Grok 4.6 pricing at $2 input and $6 output per million tokens.
- It leads the launch table on EEBench (64.0%) and Harvey Legal Agent Benchmark (19.6%).
- Fable 5.1 Max still leads on 4 of 7 benchmarks, including Terminal-Bench 4.0.
- The API offers a 500,000-token context window and 4 reasoning levels up to xhigh.
- Only 3.3% of risky dual-use prompts passed on SpaceXAI’s HackerBench v0.3.
- Grok 4.7 保持了 Grok 4.6 的定价，即每百万 token 输入 2 美元，输出 6 美元。
- 它在 EEBench（64.0%）和 Harvey Legal Agent Benchmark（19.6%）的发布榜单上名列前茅。
- Fable 5.1 Max 仍在 7 项基准测试中的 4 项保持领先，包括 Terminal-Bench 4.0。
- API 提供 500,000 token 的上下文窗口，以及最多至 xhigh 的 4 个推理级别。
- 在 SpaceXAI 的 HackerBench v0.3 上，仅有 3.3% 的高风险双重用途提示词被放行。
Check out the announcement, docs, and X post. All credit goes to the researcher of this project. Also, feel free to follow us on Twitter and don’t forget to join our 150k+ML SubReddit and Subscribe to our Newsletter. Wait! are you on telegram? now you can join us on telegram as well.
查看公告、文档和 X 帖子。所有功劳归于本项目的研究者。也欢迎在 Twitter 上关注我们，别忘了加入我们拥有 15 万+成员的 ML SubReddit 并订阅我们的新闻通讯。等等！你在 Telegram 上吗？现在你也可以在 Telegram 上加入我们。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力