- 图表数据
- artificial-analysis
- Pronunciation Robustness by Category
artificial-analysis分组柱状图 · 64 个数据点 · Gemini 3
Pronunciation Robustness by Category
artificial-analysis 发布的一张图背后的数字，用原生图表重画。对照原图，研究数据，再做你自己的。
简化重绘
由我们用下面的数字重新绘制。悬停、点按或用方向键查看。这是简化重绘：它用我们自己的样式展示数字，不是原图的样子。强调色标出这张图的主角。表格里带 ~ 的是估算值。
数字
64 个点中前 10 个免费| 分组 | 系列 | 数值 | 来源 | 置信 |
|---|---|---|---|---|
| Gemini 2.0 Flash TTS | Standalone Terms (Term Pronunciation) | ~89.5% | 估算 | 0.30 |
| Gemini 2.0 Flash TTS | Contextually Appropriate (Contextual Disambiguation) | ~88.2% | 估算 | 0.30 |
| Gemini 2.0 Flash TTS | Preserving Exact Sequences (Sequence Fidelity) | ~87.6% | 估算 | 0.30 |
| Gemini 2.0 Flash TTS | Expanding Shorthand (Text Normalization) | ~87.4% | 估算 | 0.30 |
| Gemini 2.5 Flash TTS | Standalone Terms (Term Pronunciation) | ~88.2% | 估算 | 0.30 |
| Gemini 2.5 Flash TTS | Contextually Appropriate (Contextual Disambiguation) | ~87.4% | 估算 | 0.30 |
| Gemini 2.5 Flash TTS | Preserving Exact Sequences (Sequence Fidelity) | ~87.6% | 估算 | 0.30 |
| Gemini 2.5 Flash TTS | Expanding Shorthand (Text Normalization) | ~87.4% | 估算 | 0.30 |
| SpeechX-TTS v3 | Standalone Terms (Term Pronunciation) | ~87.6% | 估算 | 0.30 |
| SpeechX-TTS v3 | Contextually Appropriate (Contextual Disambiguation) | ~87.4% | 估算 | 0.30 |
Max 数据，目前留邮箱免费看
查看全部 64 个数据点并下载文件
这张图的完整表格、JSON、CSV 和 agent skill 文件。它们属于 Max；原型阶段，留邮箱即可解锁。
| SpeechX-TTS v3 | Preserving Exact Sequences (Sequence Fidelity) | ~87.6% | 估算 | 0.30 |
|---|---|---|---|---|
| SpeechX-TTS v3 | Expanding Shorthand (Text Normalization) | ~87.4% | 估算 | 0.30 |
| Gemini 2.0 Flash TTS | Standalone Terms (Term Pronunciation) | ~87.4% | 估算 | 0.30 |
| Gemini 2.0 Flash TTS | Contextually Appropriate (Contextual Disambiguation) | ~87.4% | 估算 | 0.30 |
| Gemini 2.0 Flash TTS | Preserving Exact Sequences (Sequence Fidelity) | ~87.4% | 估算 | 0.30 |
| Gemini 2.0 Flash TTS | Expanding Shorthand (Text Normalization) | ~87.4% | 估算 | 0.30 |
| Eleven v3 | Standalone Terms (Term Pronunciation) | ~85.6% | 估算 | 0.30 |
| Eleven v3 | Contextually Appropriate (Contextual Disambiguation) | ~85.6% | 估算 | 0.30 |
| Eleven v3 | Preserving Exact Sequences (Sequence Fidelity) | ~85.6% | 估算 | 0.30 |
| Eleven v3 | Expanding Shorthand (Text Normalization) | ~85.6% | 估算 | 0.30 |
| Conversational | Standalone Terms (Term Pronunciation) | ~84.8% | 估算 | 0.30 |
| Conversational | Contextually Appropriate (Contextual Disambiguation) | ~84.8% | 估算 | 0.30 |
| Conversational | Preserving Exact Sequences (Sequence Fidelity) | ~84.8% | 估算 | 0.30 |
| Conversational | Expanding Shorthand (Text Normalization) | ~84.8% | 估算 | 0.30 |
| OpenAudio S1-TTS | Standalone Terms (Term Pronunciation) | ~81.6% | 估算 | 0.30 |
| OpenAudio S1-TTS | Contextually Appropriate (Contextual Disambiguation) | ~81.6% | 估算 | 0.30 |
| OpenAudio S1-TTS | Preserving Exact Sequences (Sequence Fidelity) | ~81.6% | 估算 | 0.30 |
| OpenAudio S1-TTS | Expanding Shorthand (Text Normalization) | ~81.6% | 估算 | 0.30 |
| RealTime TTS-2.0 | Standalone Terms (Term Pronunciation) | ~80.7% | 估算 | 0.30 |
| RealTime TTS-2.0 | Contextually Appropriate (Contextual Disambiguation) | ~80.7% | 估算 | 0.30 |
| RealTime TTS-2.0 | Preserving Exact Sequences (Sequence Fidelity) | ~80.7% | 估算 | 0.30 |
| RealTime TTS-2.0 | Expanding Shorthand (Text Normalization) | ~80.7% | 估算 | 0.30 |
| Sesame CSM-1B | Standalone Terms (Term Pronunciation) | ~77.6% | 估算 | 0.30 |
| Sesame CSM-1B | Contextually Appropriate (Contextual Disambiguation) | ~77.6% | 估算 | 0.30 |
| Sesame CSM-1B | Preserving Exact Sequences (Sequence Fidelity) | ~77.6% | 估算 | 0.30 |
| Sesame CSM-1B | Expanding Shorthand (Text Normalization) | ~77.6% | 估算 | 0.30 |
| Sesame CSM-1B | Standalone Terms (Term Pronunciation) | ~74.8% | 估算 | 0.30 |
| Sesame CSM-1B | Contextually Appropriate (Contextual Disambiguation) | ~74.8% | 估算 | 0.30 |
| Sesame CSM-1B | Preserving Exact Sequences (Sequence Fidelity) | ~74.8% | 估算 | 0.30 |
| Sesame CSM-1B | Expanding Shorthand (Text Normalization) | ~74.8% | 估算 | 0.30 |
| Sesame CSM-1B | Standalone Terms (Term Pronunciation) | ~73.6% | 估算 | 0.30 |
| Sesame CSM-1B | Contextually Appropriate (Contextual Disambiguation) | ~73.6% | 估算 | 0.30 |
| Sesame CSM-1B | Preserving Exact Sequences (Sequence Fidelity) | ~73.6% | 估算 | 0.30 |
| Sesame CSM-1B | Expanding Shorthand (Text Normalization) | ~73.6% | 估算 | 0.30 |
| Sesame CSM-1B | Standalone Terms (Term Pronunciation) | ~73.2% | 估算 | 0.30 |
| Sesame CSM-1B | Contextually Appropriate (Contextual Disambiguation) | ~73.2% | 估算 | 0.30 |
| Sesame CSM-1B | Preserving Exact Sequences (Sequence Fidelity) | ~73.2% | 估算 | 0.30 |
| Sesame CSM-1B | Expanding Shorthand (Text Normalization) | ~73.2% | 估算 | 0.30 |
| Sesame CSM-1B | Standalone Terms (Term Pronunciation) | ~72.1% | 估算 | 0.30 |
| Sesame CSM-1B | Contextually Appropriate (Contextual Disambiguation) | ~72.1% | 估算 | 0.30 |
| Sesame CSM-1B | Preserving Exact Sequences (Sequence Fidelity) | ~72.1% | 估算 | 0.30 |
| Sesame CSM-1B | Expanding Shorthand (Text Normalization) | ~72.1% | 估算 | 0.30 |
| Sesame CSM-1B | Standalone Terms (Term Pronunciation) | ~68.8% | 估算 | 0.30 |
| Sesame CSM-1B | Contextually Appropriate (Contextual Disambiguation) | ~68.8% | 估算 | 0.30 |
| Sesame CSM-1B | Preserving Exact Sequences (Sequence Fidelity) | ~68.8% | 估算 | 0.30 |
| Sesame CSM-1B | Expanding Shorthand (Text Normalization) | ~68.8% | 估算 | 0.30 |
| Sesame CSM-1B | Standalone Terms (Term Pronunciation) | ~63.6% | 估算 | 0.30 |
| Sesame CSM-1B | Contextually Appropriate (Contextual Disambiguation) | ~63.6% | 估算 | 0.30 |
| Sesame CSM-1B | Preserving Exact Sequences (Sequence Fidelity) | ~63.6% | 估算 | 0.30 |
| Sesame CSM-1B | Expanding Shorthand (Text Normalization) | ~63.6% | 估算 | 0.30 |
| Sesame CSM-1B | Standalone Terms (Term Pronunciation) | ~62.1% | 估算 | 0.30 |
| Sesame CSM-1B | Contextually Appropriate (Contextual Disambiguation) | ~62.1% | 估算 | 0.30 |
| Sesame CSM-1B | Preserving Exact Sequences (Sequence Fidelity) | ~62.1% | 估算 | 0.30 |
| Sesame CSM-1B | Expanding Shorthand (Text Normalization) | ~62.1% | 估算 | 0.30 |
下载这张图的数据
Max 订阅者可以通过 API 取同样的数据。 API 文档
API · Max
用 API 调用这张图
Max 套餐可以取这张图的数字（每个数字都带来源），格式为 JSON 或 CSV，附重画规格；也可以让智能体通过 MCP 调用。图表 id：454dae92855a。
数值是近似的。我们对这张图的两次读取只部分一致（置信度 0.30，满分 1），引用前请对照原图。API 会标记 lowConfidence，加 ?min_confidence=0.5 可以把它排除。
| 路径 | 返回 | 套餐 | |
|---|---|---|---|
| GET | /api/v1/charts/454dae92855a | 标题、发布者、日期、图型、数据点数；用 Max 密钥时附数据链接 | 开放 |
| GET | /api/v1/charts/454dae92855a/data | 每个数据点及其来源和置信度，加上重画规格。加 ?format=csv 得到 CSV | Max |
| GET | /api/v1/charts/454dae92855a/render?view=1 | 本页使用的只读绘图视图 | 开放 |
| POST | /mcp | 工具 get_chart_data 加这个 id，供 Claude、Cursor 等智能体使用 | Max |
curlMax
curl -s https://viral-chart.com/api/v1/charts/454dae92855a/data \
-H "Authorization: Bearer $VIRALCHART_KEY"Claude CodeMax
claude mcp add --transport http viralchart https://viral-chart.com/mcp \
--header "Authorization: Bearer $VIRALCHART_KEY"每个密钥每分钟 60 次，超出返回 429 和 Retry-After。每个账号最多 5 个密钥，在账户页创建，只显示一次。没有 Max 时数据接口返回 402，并附套餐链接。这些数字描述的是发布者拥有的图，每个文件都带授权说明。
做一张这样的图
这张图的形式，你的数字。
实时预览。所有内容都留在你的设备上。
选择形式
简化形式：用我们自己的样式，不是原图的样子。你的数字从这张图的数据开始。
换上你的数字
这是这张图自己的名称和数字，方便你看清形式。直接改写任何名称或数字，预览会随之重画。发布前请把每一个都换成你的。
这是这张图的数字，方便你看清形式。发布前请把每一个都换成你的。
你的品牌
强调色
Logo（可选）只在你的设备上绘制，不会上传。
画面
导出图片
免费图片的角落有一行小字 “Made with ViralChart”，别人就是通过这一行找到我们的。
Google 的更多数据
全部图表数据从图片估算相似重绘 8.3Gemini 3.8 Flash CyberGoogle · 柱状图 · 3 个数据点从图片估算Gemini 2.5 Flash Native AudioGoogle · 跑分柱状图 · 7 个数据点从图片估算CWE-bench v1 leaderboardGoogle · 柱状图 · 11 个数据点从图片估算Gemini 3 Deep ThinkGoogle · 跑分柱状图 · 18 个数据点从图片估算相似重绘 8.2Speech Reasoning: Big Bench Audioartificial-analysis · 柱状图 · 14 个数据点从图片估算Discovering securityGoogle · 跑分柱状图 · 4 个数据点
常见问题
这些数字有多准？
它们由两遍独立读图估算得出，置信度 0.30（0 到 1）。图上印着的数字读得准，从坐标轴读出的值可能有几个百分点的偏差。
这些数据可以直接用吗？
这些数字描述的是 artificial-analysis 发布的图表。图片、版式和样式归他们所有。使用数字时请注明发布者，图片请用你自己的数据来做。
怎么拿到文件？
JSON、CSV 和 agent skill 文件在本页的邮箱解锁区。原型里它们是静态文件：/api/charts/454dae92855a.json 和 /api/charts/454dae92855a.skill.md。