1. 图表数据
  2. Kimi (Moonshot AI)
  3. Evaluation Time: 2025-10-10
Kimi (Moonshot AI)表格图 · 107 个数据点 · Kimi (Moonshot AI)

Evaluation Time: 2025-10-10

Kimi (Moonshot AI) 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。

从图片估算 63%Kimi (Moonshot AI) 全部模型全部图表数据
图表来自 Kimi (Moonshot AI). 来源: x.com

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

107 个点中前 10 个免费
分组系列数值来源置信
MoonshotAICount of Finish Reason stop2679估算0.97
MoonshotAICount of Finish Tool calls1286估算0.97
MoonshotAICount of Finish others35估算0.97
MoonshotAISchema Validation Error Count0估算0.97
MoonshotAISuccessful Tool Call Count1286估算0.97
MoonshotAI TurboCount of Finish Reason stop2659估算0.97
MoonshotAI TurboCount of Finish Tool calls1301估算0.97
MoonshotAI TurboCount of Finish others40估算0.97
MoonshotAI TurboSchema Validation Error Count0估算0.97
MoonshotAI TurboSuccessful Tool Call Count1301估算0.97

Max 数据，目前留邮箱免费看

查看全部 107 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。

解锁全部 35 家巨头。每周一封，随时退订。 看看能得到什么 · 登录

MoonshotAI TurboSimilarity compared to the official Implementation99.26估算0.97
NovitaAICount of Finish Reason stop2717估算0.97
NovitaAICount of Finish Tool calls1279估算0.97
NovitaAICount of Finish others4估算0.97
NovitaAISchema Validation Error Count~26估算0.30
NovitaAISuccessful Tool Call Count~1279估算0.30
NovitaAISimilarity compared to the official Implementation~92.87估算0.30
GroqCount of Finish Reason stop2739估算0.97
GroqCount of Finish Tool calls989估算0.97
GroqCount of Finish others~4估算0.30
GroqSchema Validation Error Count~26估算0.30
GroqSuccessful Tool Call Count989估算0.97
GroqSimilarity compared to the official Implementation~89.39估算0.30
FireworksCount of Finish Reason stop2546估算0.97
FireworksCount of Finish Tool calls1443估算0.97
FireworksCount of Finish others8估算0.97
FireworksSchema Validation Error Count~2估算0.30
FireworksSuccessful Tool Call Count~1443估算0.30
FireworksSimilarity compared to the official Implementation88.83估算0.97
BasetenCount of Finish Reason stop~2598估算0.50
BasetenCount of Finish Tool calls~1396估算0.50
BasetenCount of Finish others~6估算0.50
BasetenSchema Validation Error Count~2估算0.50
BasetenSuccessful Tool Call Count~1396估算0.50
BasetenSimilarity compared to the official Implementation~88.57估算0.50
DeepInfraCount of Finish Reason stop2552估算0.97
DeepInfraCount of Finish Tool calls~1446估算0.30
DeepInfraCount of Finish others~2估算0.30
DeepInfraSchema Validation Error Count~2估算0.30
DeepInfraSuccessful Tool Call Count~1446估算0.30
DeepInfraSimilarity compared to the official Implementation88.05估算0.97
VolcCount of Finish Reason stop2572估算0.97
VolcCount of Finish Tool calls~1368估算0.30
VolcCount of Finish others60估算0.97
VolcSchema Validation Error Count~1估算0.30
VolcSuccessful Tool Call Count~1368估算0.30
VolcSimilarity compared to the official Implementation87.595估算0.97
TogetherCount of Finish Reason stop2725估算0.97
TogetherCount of Finish Tool calls1271估算0.97
TogetherCount of Finish others1估算0.97
TogetherSchema Validation Error Count~1估算0.30
TogetherSuccessful Tool Call Count~1271估算0.30
TogetherSimilarity compared to the official Implementation86.6估算0.97
SiliconFlowCount of Finish Reason stop2626估算0.97
SiliconFlowCount of Finish Tool calls985估算0.97
SiliconFlowCount of Finish others389估算0.97
SiliconFlowSchema Validation Error Count1估算0.97
SiliconFlowSuccessful Tool Call Count984.5估算0.97
SiliconFlowSimilarity compared to the official Implementation86.08估算0.97
NebiusCount of Finish Reason stop2729估算0.97
NebiusCount of Finish Tool calls~633估算0.30
NebiusCount of Finish others37估算0.97
NebiusSchema Validation Error Count2估算0.97
NebiusSuccessful Tool Call Count~633估算0.30
NebiusSimilarity compared to the official Implementation~82.17估算0.30
InfiniGenCount of Finish Reason stop~3327估算0.50
InfiniGenCount of Finish Tool calls~860估算0.50
InfiniGenCount of Finish others~411估算0.50
InfiniGenSchema Validation Error Count~2估算0.50
InfiniGenSuccessful Tool Call Count~860估算0.50
InfiniGenSimilarity compared to the official Implementation~70.49估算0.50
ChutesCount of Finish Reason stop~3866估算0.30
ChutesCount of Finish Tool calls~131估算0.30
ChutesCount of Finish others~0估算0.30
ChutesSchema Validation Error Count~2估算0.30
ChutesSuccessful Tool Call Count~131估算0.30
ChutesSimilarity compared to the official Implementation~49.12估算0.30
AtlasCloudCount of Finish Reason stop3866.5估算0.97
AtlasCloudCount of Finish Tool calls~128估算0.30
AtlasCloudCount of Finish others0估算0.97
AtlasCloudSchema Validation Error Count~2估算0.30
AtlasCloudSuccessful Tool Call Count~128估算0.30
AtlasCloudSimilarity compared to the official Implementation48.93估算0.97
MoonshotAIModel Name~0估算0.50
Moonshot AI TurboModel Name~0估算0.50
NovitaAIModel Name~0估算0.50
GroqModel Name~0估算0.50
FireworksModel Name~0估算0.50
DeepinfraModel Name~0估算0.50
VolcModel Name~0估算0.50
TogetherModel Name~0估算0.50
SiliconFlowModel Name~0估算0.50
NebiusModel Name~0估算0.50
ChutesModel Name~0估算0.50
AtlasCloudModel Name~0估算0.50
MoonshotAIProviders~0估算0.50
Moonshot AI TurboProviders~0估算0.50
NovitaAIProviders~0估算0.50
GroqProviders~0估算0.50
FireworksProviders~0估算0.50
DeepinfraProviders~0估算0.50
VolcProviders~0估算0.50
TogetherProviders~0估算0.50
SiliconFlowProviders~0估算0.50
NebiusProviders~0估算0.50
ChutesProviders~0估算0.50
AtlasCloudProviders~0估算0.50
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

JSONagent skill 文件

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：c645ee4f5709。

查看 Max 套餐API 文档
路径返回套餐
GET/api/v1/charts/c645ee4f5709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/c645ee4f5709/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 CSVMax
GET/api/v1/charts/c645ee4f5709/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/c645ee4f5709/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。

Kimi (Moonshot AI) 的更多数据

全部图表数据
从图片估算相似重绘 9.0Model Performance ComparisonKimi (Moonshot AI) · 跑分柱状图 · 15 个数据点从图片估算Model BenchmarksKimi (Moonshot AI) · 跑分柱状图 · 47 个数据点从图片估算相似重绘 7.9Coding and Agents BenchmarksKimi (Moonshot AI) · 跑分柱状图 · 24 个数据点从图片估算Model Performance ComparisonKimi (Moonshot AI) · 跑分柱状图 · 38 个数据点从图片估算Internal Knowledge Work BenchKimi (Moonshot AI) · 跑分柱状图 · 9 个数据点从图片估算相似重绘 6.3Frontend Code Kimi-K3: Ranked #1arena · 柱状图 · 23 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 0.63（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 Kimi (Moonshot AI) 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/c645ee4f5709.json 和 /api/charts/c645ee4f5709.skill.md。