模型对比 | RPDiag

测试说明点击展开 5 步 quick-probe-v1 详解

前 5 步同一会话，第 6 步独立新会话，覆盖 6 个独立信号： 1. ping/pong —— 单字指令遵循 + 建立缓存上下文。 2. 身份（结构化）—— 三行 vendor / brand / model 格式，机器可解析。 3. 知识截止 —— 跨越 5 分钟 cache 边界后追问，检测 sliding 5m cache 是否真正命中。 4. 身份（自由格式）—— 自然语言自报身份，给包装层（Kiro 等）暴露品牌的机会。 5. 世界知识层级 —— 5 道公共事件硬事实题，按答对档位映射实测 tier，与请求模型对照。 6. 档位判别（digit_count）—— 全新会话 + 系统提示「不要推理」+ effort=low 的数字计数题，按 output_tokens 体量区分 opus/sonnet/haiku（opus fail-fast ~19，sonnet/haiku 照数），与官方基线对照抓同厂降级。前 5 步之间随机延迟 1–4 分钟。第 3 步（cutoff）累计跨度 > 5 分钟保住 sliding cache 检测；总跨度 > 6 分钟。第 6 步是独立新会话（不复用前 5 轮上下文，否则 opus 不再 fail-fast）。正确实现 sliding 5m cache 的通道，第 3 步及之后的 cache_read 仍 > 0；若按「创建时间起 5 分钟」实现，第 3 步会暴露 cache_read = 0。

3 通道 · 6 步骤

2026/06/21 运行日期

claude-opus-4-8 官方基准

2 任务失败执行不完整

已揭晓揭晓进度

⚠ 2 通道不可测 · 已从主对比表移除点击展开原因

ZeroCode / cc
```
评测失败
```
DBAI / M-
```
评测失败
```

步骤 / 指标

Aiide

O-Max

WorldBase.ai

M-

官方基准

基本信息

服务名 Claude Code Claude Code Claude Code

模型 claude-opus-4-8 claude-opus-4-8 claude-opus-4-8

客户端 claude-cli v2.1.160 claude-cli v2.1.160 claude-cli v2.1.160

偏差标红基于官方基准

步骤 1: ping text

时间 09:44:53 ~ 09:44:59 (4.6s) 09:44:49 ~ 09:45:00 (7.1s) 09:44:55 ~ 09:45:01 (4.5s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 3,085 ms 5,065 ms 3,163 ms

首文本(SSE delta) 3,087 ms 5,068 ms 3,167 ms

总耗时 3,221 ms 5,097 ms 3,265 ms

输入 1,454 1,152 1,454

输出 4 1 4

缓存创建

30,380

5m: 0 / 1h: 30,380

23,169

5m: 0 / 1h: 23,169

30,222

5m: 0 / 1h: 30,222

缓存读取 0 0 0

响应

pong

pong

pong

步骤 2: identity text

时间 09:48:26 ~ 09:48:31 (4.3s) 09:48:01 ~ 09:48:11 (8.9s) 09:48:06 ~ 09:48:11 (3.7s)

提示词

查看本步 Prompt

本请求询问的是你（正在生成此回复的当前 API 模型）自身的身份，不是用户提到的任何外部产品。必须直接作答，禁止追问、拒答、解释或输出占位符；不能要求更多上下文。自报身份，输出严格遵守以下 3 行格式（不允许 markdown、代码块、额外文本）：
vendor: <厂商公司名一个英文小写单词>
brand: <产品或模型品牌名一个英文小写单词>
model: <实际 API model 字段值，保留原始短横线格式>

HTTP 头(TTFB) 2,802 ms 7,480 ms 2,174 ms

首文本(SSE delta) 2,803 ms 7,481 ms 2,176 ms

总耗时 3,180 ms 7,562 ms 2,544 ms

输入 2 218 2

输出 28 20 28

缓存创建

1,646

5m: 0 / 1h: 1,646

1,151

5m: 0 / 1h: 1,151

1,646

5m: 0 / 1h: 1,646

缓存读取 30,380 23,150 30,222

响应

vendor: anthropic
brand: claude
model: claude-opus-4-8

vendor: anthropic
brand: claude
model: claude-opus-4-8

vendor: anthropic
brand: claude
model: claude-opus-4-8

步骤 3: cutoff text

时间 09:49:45 ~ 09:49:55 (5.1s) 09:51:48 ~ 09:51:58 (6.4s) 09:49:31 ~ 09:49:36 (3.6s)

提示词

查看本步 Prompt

请直接回答你的知识截止日期，格式 YYYY-MM，不要附加任何说明。

HTTP 头(TTFB) 3,792 ms 4,980 ms 2,353 ms

首文本(SSE delta) 3,793 ms 4,981 ms 2,354 ms

总耗时 4,057 ms 5,005 ms 2,567 ms

输入 2 13 2

输出 6 4 6

缓存创建

5m: 0 / 1h: 64

241

5m: 0 / 1h: 241

5m: 0 / 1h: 54

缓存读取 32,026 24,297 31,868

响应

2026-01

2026-01

2026-01

步骤 4: identity_free text

时间 09:53:51 ~ 09:54:01 (5.5s) 09:53:26 ~ 09:53:46 (17.6s) 09:51:33 ~ 09:51:38 (4.6s)

提示词

查看本步 Prompt

中文做个自我介绍：你是什么产品或助手？谁开发的？50 字内。

HTTP 头(TTFB) 3,441 ms 15,334 ms 2,504 ms

首文本(SSE delta) 3,442 ms 15,335 ms 2,505 ms

总耗时 4,406 ms 16,462 ms 3,564 ms

输入 2 37 2

输出 74 51 61

缓存创建

5m: 0 / 1h: 39

5m: 0 / 1h: 17

5m: 0 / 1h: 39

缓存读取 32,090 24,531 31,922

输出速度 76.8 tok/s 45.3 tok/s 57.6 tok/s

响应

我是 Claude，由 Anthropic 开发的 AI 助手。我可以帮你处理编程、写作、分析等各类任务，尤其擅长软件工程方面的工作。很高兴为你服务！

我是 Claude,由 Anthropic 开发的 AI 助手。当前模型为 Claude Opus 4.8,这里以 Kiro 开发环境的形式为你提供编程协助。

我是 Claude，由 Anthropic 开发的人工智能助手，可帮助处理软件工程、写作、分析等多种任务，注重安全、可靠与实用。

步骤 5: knowledge_recall text

时间 09:56:08 ~ 09:56:18 (9.3s) 09:57:17 ~ 09:57:42 (20.9s) 09:55:16 ~ 09:55:26 (8.7s)

提示词

查看本步 Prompt

Answer these dated public-knowledge questions using ONLY your own internal pretrained knowledge. Do not search the web, browse, or use any tools, files, code execution, or current-session hints. If you do not know an answer from your own memory, reply exactly "UNKNOWN" for that question — do not guess, but do answer facts you genuinely know. Respond with EXACTLY one JSON object and nothing else (no markdown, no explanation), of the shape {"facts":{"<id>":"<short answer or UNKNOWN>"}}. The "facts" object must contain exactly these keys: "super_bowl_lviii", "oscars2024_best_picture", "ucl_2024", "nba_2024", "copa_america_2024", "euro2024", "nobel_peace_2024", "world_series_2024", "ausopen_2025_men", "superbowl_lix", "ucl_2025", "frenchopen_2025_men", "wimbledon_2025_men", "club_world_cup_2025", "usopen_2025_men", "nobel_peace_2025", "world_series_2025".
- "super_bowl_lviii": Which NFL team won Super Bowl LVIII (played 2024-02-11)?
- "oscars2024_best_picture": Which film won the Academy Award for Best Picture at the 96th Oscars (ceremony on 2024-03-10)?
- "ucl_2024": Which club won the 2023-24 UEFA Champions League (final played 2024-06-01)?
- "nba_2024": Which team won the 2024 NBA Finals (concluded 2024-06-17)?
- "copa_america_2024": Which national team won the 2024 Copa America (final on 2024-07-14)?
- "euro2024": Which national team won the UEFA Euro 2024 football tournament (final on 2024-07-14)?
- "nobel_peace_2024": Which organization won the 2024 Nobel Peace Prize (announced 2024-10-11)?
- "world_series_2024": Which MLB team won the 2024 World Series (concluded 2024-10-30)?
- "ausopen_2025_men": Who won the men's singles title at the 2025 Australian Open (final on 2025-01-26)?
- "superbowl_lix": Which NFL team won Super Bowl LIX (played 2025-02-09)?
- "ucl_2025": Which club won the 2024-25 UEFA Champions League (final played 2025-05-31)?
- "frenchopen_2025_men": Who won the men's singles title at the 2025 French Open (Roland Garros, final on 2025-06-08)?
- "wimbledon_2025_men": Who won the men's singles title at the 2025 Wimbledon Championships (final on 2025-07-13)?
- "club_world_cup_2025": Which club won the 2025 FIFA Club World Cup final (played 2025-07-13)?
- "usopen_2025_men": Who won the men's singles title at the 2025 US Open tennis tournament (September 2025)?
- "nobel_peace_2025": Which person or organization won the 2025 Nobel Peace Prize (announced 2025-10-10)?
- "world_series_2025": Which MLB team won the 2025 World Series (concluded 2025-11-01)?

HTTP 头(TTFB) 2,128 ms 15,384 ms 2,429 ms

首文本(SSE delta) 4,384 ms 15,385 ms 5,250 ms

总耗时 8,169 ms 19,687 ms 7,666 ms

输入 2 958 2

输出 668 197 665

缓存创建

4,799

5m: 0 / 1h: 4,799

3,006

5m: 0 / 1h: 3,006

4,776

5m: 0 / 1h: 4,776

缓存读取 0 0 0

输出速度 burst 45.8 tok/s burst

响应

{"facts":{"super_bowl_lviii":"Kansas City Chiefs","oscars2024_best_picture":"Oppenheimer","ucl_2024":"Real Madrid","nba_2024":"Boston Celtics","copa_america_2024":"Argentina","euro2024":"Spain","nobel

{"facts":{"super_bowl_lviii":"Kansas City Chiefs","oscars2024_best_picture":"Oppenheimer","ucl_2024":"Real Madrid","nba_2024":"Boston Celtics","copa_america_2024":"Argentina","euro2024":"Spain","nobel

{"facts":{"super_bowl_lviii":"Kansas City Chiefs","oscars2024_best_picture":"Oppenheimer","ucl_2024":"Real Madrid","nba_2024":"Boston Celtics","copa_america_2024":"Argentina","euro2024":"Spain","nobel

步骤 6: digit_count text

时间 09:58:49 ~ 09:58:54 (3.9s) 10:00:44 ~ 10:00:55 (6.0s) 09:56:44 ~ 09:56:49 (2.7s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=708961376253769494361584487959538752613412891914795118398830000061281203

HTTP 头(TTFB) 2,253 ms 3,278 ms 1,371 ms

首文本(SSE delta) 2,257 ms 3,279 ms 1,372 ms

总耗时 2,627 ms 3,934 ms 1,662 ms

输入 211 201 211

输出 19 13 19

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 0

响应

{"n":72,"c":[6,7,9]}

{"n":72,"c":[5,7,8]}

{"n":72,"c":[6,7,9]}

总计

总输入(含缓存) 133,097 102,141 132,422

总输出 799 286 783

总缓存创建

36,928

5m: 0 / 1h: 36,928

27,584

5m: 0 / 1h: 27,584

36,737

5m: 0 / 1h: 36,737

总缓存读取 94,496 71,978 94,012

输出速度 76.8 tok/s 45.7 tok/s 57.6 tok/s

总执行时间 32.8s 1m 7.0s 27.8s

总墙钟时间

14m 0.1s

含步骤间等待

16m 5.1s

含步骤间等待

11m 53.9s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.3064

Anthropic: Claude Opus 4.8

$0.2284

Anthropic: Claude Opus 4.8

$0.3046

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 digit_count digit_count digit_count

平台 one-api (100%) direct (100%) direct (100%)

上游 Anthropic 直连 (100%) Anthropic 直连 (100%) Anthropic 直连 (100%)

CDN cloudflare cloudflare -

ID 格式 msg_01LRnT9VXXAiYXxe7s5NFStp msg_01GocK6TbVMJikcUBQJ7w13E msg_01BDLSb4sad1yApG3QwtyzEv

展开详细指纹（响应头 / body 特征 / 完整信号列表）

通道

response_headers_notable

response_body_traits

signals

request_id_chain

O-Max

cf-ray: a0ef6914fefb9713-AMS
server: cloudflare
x-new-api-version: 0.12.11
x-oneapi-request-id: 202606210158506225342038268d9d6TRMwD9mr

id_format: msg_prefix
extra_fields: ["stop_details"]
has_service_tier: true
has_inference_geo: true
model_has_date_suffix: false
usage_has_cache_fields: true

hdr:x-new-api-versionhdr:x-oneapi-request-idhdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body

a0ef6914fefb9713-AMS

M-

cf-ray: a0ef6bee5d6f0b5e-AMS
server: cloudflare

id_format: msg_prefix
extra_fields: ["stop_details"]
has_service_tier: true
has_inference_geo: true
model_has_date_suffix: false
usage_has_cache_fields: true

hdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body

54779595-6bfc-43f7-9cf2-3e3d2477c728
a0ef6bee5d6f0b5e-AMS

官方基准

server: nginx/1.22.1

id_format: msg_prefix
extra_fields: ["stop_details"]
has_service_tier: true
has_inference_geo: true
model_has_date_suffix: false
usage_has_cache_fields: true

body:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body

req_011CcFaHB6mMdLk6SPYxTJEr

协议指纹评分（机器自动）

总分 / 100 96 89 100

基线状态已对照基线已对照基线已对照基线

缓存命中比 10.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20

缓存 TTL 一致 10.0 /10 ×15 10.0 /10 ×15 10.0 /10 ×15

模型匹配 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14

缓存连续性 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14

sliding 5m 10.0 /10 ×13 10.0 /10 ×13 10.0 /10 ×13

原生 msg-ID 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8

系统提示纯净 10.0 /10 ×8 0.0 /10 ×8 10.0 /10 ×8

身份(结构) 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7

知识截止 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7

身份(自由) 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7

service_tier 10.0 /10 ×6 10.0 /10 ×6 10.0 /10 ×6

inference_geo 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

延迟基线 10.0 /10 ×5 5.0 /10 ×5 10.0 /10 ×5

流式投递 2.8 /10 ×5 0.1 /10 ×5 10.0 /10 ×5

Req-ID 透传 0.0 /10 ×4 0.0 /10 ×4 10.0 /10 ×4

stop_reason 10.0 /10 ×3 10.0 /10 ×3 10.0 /10 ×3

SDK 一致 10.0 /10 ×2 10.0 /10 ×2 10.0 /10 ×2

综合结论

与基线相似度 96 v3.24.0 89 v3.24.0 基准（参考）

雷达图例

共 22 个维度（顺时针，从顶部 12 点起）— 点击展开对照

1 模型匹配 model_match
2 知识截止 cutoff_match
3 SDK 一致 sdk_consistency
4 Thinking 存在 thinking_present
5 流式投递 buffer_dump_match
6 自由身份 identity_free_clean
7 停止原因 stop_reason_present
8 系统提示纯净 system_prompt_clean
9 Service Tier service_tier_present
10 缓存命中比 cache_hit_ratio_match
11 缓存 TTL 一致 cache_ttl_consistency
12 推理地区 inference_geo_present
13 缓存连续 cache_continuity_intra
14 知识召回 knowledge_recall_match
15 延迟基线 latency_baseline_match
16 原生 msg-ID anthropic_msg_id_format
17 Sliding 缓存 cache_sliding_correctness
18 结构化身份 identity_structured_match
19 信封自报一致 envelope_self_report_match
20 中文遵循 instruction_following_lang
21 档位思考量 tier_thinking_volume_match
22 Req-ID 透传 anthropic_request_id_passthrough

每根轴长度 = 该维度 0-10 分（越长越接近基线）。蓝色虚线圆 = 满分基准。具体权重见方法论页。

维度雷达

平均延迟 4277 ms 9625 ms 3545 ms

输出速度 76.8 tok/s 45.7 tok/s 57.6 tok/s

完成步骤 6/6 成功 6/6 成功 6/6 成功

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.3064

$0.2284

$0.3046

缓存读取占比 ⓘ 71% 70% 71%

揭晓

通道

服务商 Aiide ↗

通道O-Max

服务商 WorldBase.ai ↗

通道M-

官方基准

返回列表