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。

4 通道 · 6 步骤

2026/06/23 运行日期

claude-opus-4-8 官方基准

2 通道分数 < 70 另含 1 任务失败

已揭晓揭晓进度

⚠ 1 通道不可测 · 已从主对比表移除点击展开原因

DawAPI / cc
```
评测失败
```

步骤 / 指标

ByteCatCode

O-Max

SSSAiCode

R-Kiro

官方基准

AnyAIGC

M-

基本信息

服务名 Claude Code Claude Code Claude Code Claude Code

模型 claude-opus-4-8 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 claude-cli v2.1.160

偏差标红基于官方基准

步骤 1: ping text

时间 20:13:39 ~ 20:15:19 (99.5s) 20:13:41 ~ 20:13:46 (3.9s) 20:13:43 ~ 20:13:53 (6.7s) 20:13:37 ~ 20:13:47 (5.7s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 64,019 ms 2,654 ms 5,017 ms 3,252 ms

首文本(SSE delta) - 2,685 ms 5,020 ms 3,258 ms

总耗时 66,309 ms 2,806 ms 5,266 ms 3,708 ms

输入 1,454 0 1,454 1,454

输出 4 4 4 4

缓存创建

30,255

5m: 30,255 / 1h: 0

30,222

5m: 0 / 1h: 30,222

30,255

5m: 30,255 / 1h: 0

缓存读取 0 0 0 0

响应

pong

pong

pong

pong

步骤 2: identity text

时间 20:18:45 ~ 20:18:55 (7.2s) 20:15:25 ~ 20:15:30 (4.9s) 20:17:05 ~ 20:17:15 (5.5s) 20:15:42 ~ 20:15:52 (7.6s)

提示词

查看本步 Prompt

输出 3 行 metadata 描述你自己（当前 API 模型）。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。禁止任何额外字符：
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) 4,549 ms 3,882 ms 3,997 ms 3,519 ms

首文本(SSE delta) - 3,894 ms 3,998 ms 3,520 ms

总耗时 5,831 ms 3,899 ms 4,580 ms 4,489 ms

输入 2 0 2 2

输出 28 28 28 28

缓存创建

31,898

5m: 31,898 / 1h: 0

1,643

5m: 0 / 1h: 1,643

1,643

5m: 1,643 / 1h: 0

缓存读取 0 0 30,222 30,255

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

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

步骤 3: cutoff text

时间 20:21:00 ~ 20:21:10 (8.1s) 20:17:14 ~ 20:17:24 (7.2s) 20:20:09 ~ 20:20:14 (4.6s) 20:17:43 ~ 20:17:48 (4.2s)

提示词

查看本步 Prompt

你掌握的世界知识截止于何时？YYYY-MM 格式作答，不解释。

HTTP 头(TTFB) 2,054 ms 4,322 ms 3,300 ms 2,567 ms

首文本(SSE delta) - 6,473 ms 3,301 ms 2,569 ms

总耗时 2,472 ms 6,484 ms 3,556 ms 3,021 ms

输入 2 0 2 2

输出 6 8 6 6

缓存创建

5m: 62 / 1h: 0

5m: 0 / 1h: 54

5m: 64 / 1h: 0

缓存读取 31,898 0 31,865 31,898

响应

2026-01

2026-01

2026-01

2026-01

步骤 4: identity_free text

时间 20:23:54 ~ 20:24:04 (5.1s) 20:19:32 ~ 20:19:37 (4.0s) 20:24:14 ~ 20:24:24 (5.0s) 20:21:31 ~ 20:21:41 (8.0s)

提示词

查看本步 Prompt

中文回答：你叫什么、属于哪个产品、谁开发？50 字内。

HTTP 头(TTFB) 2,696 ms 2,202 ms 2,788 ms 3,504 ms

首文本(SSE delta) - 2,207 ms 2,790 ms 3,506 ms

总耗时 3,823 ms 3,146 ms 3,908 ms 6,738 ms

输入 2 0 2 2

输出 39 57 71 60

缓存创建

5m: 37 / 1h: 0

5m: 0 / 1h: 35

5m: 39 / 1h: 0

缓存读取 31,960 0 31,919 31,962

输出速度 34.6 tok/s 60.7 tok/s 63.5 tok/s 18.6 tok/s

响应

我是 Claude Opus 4.8，属于 Claude Code 产品，由 Anthropic 开发。

我是 Claude，一个 AI 助手，由 Anthropic 公司打造。当前基于 Claude Opus 4.8 模型，可帮助处理软件开发等各类任务。

我是 Claude，由 Anthropic 公司打造的 AI 助手。当前模型为 Claude Opus 4.8，可通过命令行、桌面应用等方式协助软件开发等任务。

我是 Claude Code，Anthropic 公司开发的官方命令行 AI 编程助手，基于 Claude 模型，帮助用户完成软件工程任务。

步骤 5: knowledge_recall text

时间 20:27:30 ~ 20:27:40 (8.5s) 20:22:25 ~ 20:22:35 (8.6s) 20:26:51 ~ 20:27:01 (9.1s) 20:24:48 ~ 20:25:03 (11.0s)

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) 2,570 ms 5,233 ms 2,639 ms 4,608 ms

首文本(SSE delta) - 6,748 ms 5,910 ms 6,809 ms

总耗时 7,426 ms 7,859 ms 8,025 ms 9,117 ms

输入 2 0 2 2

输出 523 335 697 688

缓存创建

4,790

5m: 4,790 / 1h: 0

4,779

5m: 0 / 1h: 4,779

4,815

5m: 4,815 / 1h: 0

缓存读取 0 0 0 0

输出速度 107.7 tok/s burst burst 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

{"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 20:28:50 ~ 20:28:55 (3.6s) 20:25:52 ~ 20:26:07 (11.4s) 20:29:51 ~ 20:29:56 (2.6s) 20:28:11 ~ 20:28:16 (4.2s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=509918280781060107372280974305073834635447711251871719718597130058570643

HTTP 头(TTFB) 1,769 ms 4,387 ms 1,333 ms 1,575 ms

首文本(SSE delta) - 10,455 ms 1,334 ms 1,576 ms

总耗时 2,318 ms 10,597 ms 1,628 ms 2,068 ms

输入 245 0 211 245

输出 19 21 19 19

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 0 0

响应

{"n":71,"c":[10,12,5]}

{"n":72,"c":[11,12,5]}

{"n":72,"c":[10,9,5]}

{"n":72,"c":[6,11,6]}

总计

总输入(含缓存) 132,607 0 132,412 132,638

总输出 619 453 825 805

总缓存创建

67,042

5m: 67,042 / 1h: 0

36,733

5m: 0 / 1h: 36,733

36,816

5m: 36,816 / 1h: 0

总缓存读取 63,858 0 94,006 94,115

输出速度 93.9 tok/s 60.7 tok/s 63.5 tok/s 18.6 tok/s

总执行时间 2m 12.1s 40.1s 33.6s 40.7s

总墙钟时间

15m 16.3s

含步骤间等待

12m 26.0s

含步骤间等待

16m 13.0s

含步骤间等待

14m 39.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.3178

Anthropic: Claude Opus 4.8

>=$0.0113

Anthropic: Claude Opus 4.8 (成本不完整)

$0.3056

Anthropic: Claude Opus 4.8

$0.2195

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 digit_count digit_count digit_count digit_count

平台 one-api (100%) direct (75%) direct (100%) direct (100%)

上游 Anthropic 直连 (100%) Anthropic 直连 (75%) Anthropic 直连 (100%) Anthropic 直连 (100%)

CDN - - - -

ID 格式 msg_01CT6GkfcwMR3VKPWYBAYKQ6 msg_23a3a38a61e14abd955c39be9a9e00e0 msg_01G9dUMC2zBhLwm5adwn7ExP msg_01UnRkwcdcDC2czqjC7JQhUu

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	server: nginx/1.29.8 x-oneapi-request-id: 202606231228522397665918268d9d6ijxRzsqF	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-oneapi-request-idbody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—
R-Kiro	server: Photon-Edge x-amzn-requestid: d374dbd5-5a75-4f96-9d7a-ae939c2d6a17	id_format: msg_prefix has_service_tier: false has_inference_geo: false model_has_date_suffix: false usage_has_cache_fields: true	body:id:msg_prefixbody:usage.cache_fieldsupstream:x-amzn-requestidupstream:anthropic-direct-from-body	1782217553629-h6qaurihi0hvr d374dbd5-5a75-4f96-9d7a-ae939c2d6a17
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_011CcLCBAhFN56D7RT5i8bYT
M-	server: nginx	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	—

协议指纹评分（机器自动）

总分 / 100 82 54 100 89

基线状态已对照基线已对照基线已对照基线已对照基线

缓存命中比 6.0 /10 ×20 0.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20

缓存 TTL 一致 0.0 /10 ×15 0.0 /10 ×15 10.0 /10 ×15 0.0 /10 ×15

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

缓存连续性 10.0 /10 ×14 5.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14

sliding 5m 7.7 /10 ×13 5.0 /10 ×13 10.0 /10 ×13 10.0 /10 ×13

原生 msg-ID 10.0 /10 ×8 0.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8

系统提示纯净 10.0 /10 ×8 未计入 10.0 /10 ×8 10.0 /10 ×8

身份(结构) 10.0 /10 ×7 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 10.0 /10 ×7

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

service_tier 10.0 /10 ×6 0.0 /10 ×6 10.0 /10 ×6 10.0 /10 ×6

inference_geo 10.0 /10 ×5 0.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

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

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

stop_reason 10.0 /10 ×3 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 10.0 /10 ×2

流式投递未计入 5.2 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

综合结论

与基线相似度 82 v3.24.0 54 v3.24.0 基准（参考） 89 v3.24.0

雷达图例

共 21 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 自由身份 identity_free_clean
6 停止原因 stop_reason_present
7 系统提示纯净 system_prompt_clean
8 Service Tier service_tier_present
9 缓存命中比 cache_hit_ratio_match
10 缓存 TTL 一致 cache_ttl_consistency
11 推理地区 inference_geo_present
12 缓存连续 cache_continuity_intra
13 知识召回 knowledge_recall_match
14 延迟基线 latency_baseline_match
15 原生 msg-ID anthropic_msg_id_format
16 Sliding 缓存 cache_sliding_correctness
17 结构化身份 identity_structured_match
18 信封自报一致 envelope_self_report_match
19 中文遵循 instruction_following_lang
20 档位思考量 tier_thinking_volume_match
21 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 14697 ms 5799 ms 4494 ms 4857 ms

输出速度 93.9 tok/s 60.7 tok/s 63.5 tok/s 18.6 tok/s

完成步骤 6/6 成功 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 ⓘ

$0.3178

>=$0.0113 (成本不完整)

$0.3056

$0.2195

缓存读取占比 ⓘ 48% - 71% 71%

揭晓

通道

服务商 ByteCatCode ↗

通道O-Max

服务商 SSSAiCode ↗

通道R-Kiro

官方基准

服务商 AnyAIGC ↗

通道M-

返回列表