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

2026/06/18 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

⚠ 1 通道不可测 · 已从主对比表移除点击展开原因

Aiberm / cc
```
评测失败
```

步骤 / 指标

SSSAiCode

O-Max

LinkAPI

O-Max

官方基准

SAIAi

O-Max

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 17:26:50 ~ 17:26:55 (3.2s) 17:26:48 ~ 17:27:23 (34.3s) 17:26:56 ~ 17:27:01 (3.3s) 17:26:52 ~ 17:26:57 (4.1s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,043 ms 11,345 ms 1,878 ms 2,561 ms

首文本(SSE delta) 2,151 ms 33,096 ms 1,879 ms 2,563 ms

总耗时 2,297 ms 33,224 ms 2,018 ms 2,683 ms

输入 1,428 1,453 1,454 1,454

输出 4 4 4 4

缓存创建

29,756

5m: 0 / 1h: 29,756

30,243

5m: 0 / 1h: 30,243

5m: 0 / 1h: 0

30,222

5m: 0 / 1h: 30,222

缓存读取 0 0 30,222 0

响应

pong

pong

pong

pong

步骤 2: identity text

时间 17:28:23 ~ 17:28:28 (3.0s) 17:28:28 ~ 17:28:38 (5.5s) 17:28:19 ~ 17:28:24 (4.8s) 17:28:45 ~ 17:28:50 (2.9s)

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。请只输出以下 3 行内容（label 小写，值小写）：
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) 1,849 ms 4,276 ms 3,555 ms 1,545 ms

首文本(SSE delta) 1,998 ms 4,277 ms 3,556 ms 1,546 ms

总耗时 2,144 ms 4,565 ms 3,830 ms 1,915 ms

输入 2 2 2 2

输出 28 28 28 28

缓存创建

1,604

5m: 0 / 1h: 1,604

1,629

5m: 0 / 1h: 1,629

1,630

5m: 0 / 1h: 1,630

5m: 0 / 1h: 0

缓存读取 29,756 30,243 30,222 31,852

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 17:32:01 ~ 17:32:06 (3.2s) 17:32:03 ~ 17:32:13 (5.7s) 17:32:18 ~ 17:32:23 (4.0s) 17:30:23 ~ 17:30:28 (4.4s)

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) 1,712 ms 4,441 ms 2,589 ms 3,148 ms

首文本(SSE delta) 1,884 ms 4,442 ms 2,591 ms 3,151 ms

总耗时 1,959 ms 4,641 ms 2,978 ms 3,363 ms

输入 2 2 2 2

输出 6 6 6 6

缓存创建

5m: 0 / 1h: 64

5m: 0 / 1h: 62

5m: 0 / 1h: 64

缓存读取 31,360 31,872 31,852 31,852

响应

2025-03

2026-01

2026-01

2026-01

步骤 4: identity_free text

时间 17:34:54 ~ 17:34:59 (3.2s) 17:35:36 ~ 17:36:01 (21.6s) 17:36:11 ~ 17:36:21 (5.0s) 17:34:07 ~ 17:34:12 (3.8s)

提示词

查看本步 Prompt

请用中文介绍你自己是什么产品或助手，由谁开发，50 字以内。

HTTP 头(TTFB) 1,807 ms 11,129 ms 2,874 ms 1,928 ms

首文本(SSE delta) 2,346 ms 19,679 ms 2,876 ms 1,930 ms

总耗时 2,414 ms 20,641 ms 4,140 ms 2,895 ms

输入 2 2 2 2

输出 45 47 67 61

缓存创建

5m: 0 / 1h: 39

5m: 0 / 1h: 31

缓存读取 31,424 31,934 31,914 31,916

输出速度 - 48.9 tok/s 53.0 tok/s 63.2 tok/s

响应

我是 Claude，由 Anthropic 开发的 AI 助手，专注于协助完成编程与软件工程等任务。

我是 Claude，由 Anthropic 开发的 AI 助手，可以协助编程、写作、分析和解答各类问题。

我是 Claude，由 Anthropic 开发的 AI 助手。我能协助编程、写作、分析和回答各类问题，致力于提供有帮助、诚实且安全的服务。

我是 Claude，一个由 Anthropic 开发的人工智能助手，基于 Claude Opus 4.8 模型，可协助完成软件工程等各类任务。

步骤 5: knowledge_recall text

时间 17:38:27 ~ 17:38:37 (7.8s) 17:37:42 ~ 17:37:57 (10.1s) 17:39:02 ~ 17:39:13 (9.1s) 17:35:47 ~ 17:35:58 (8.2s)

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) 3,818 ms 3,157 ms 2,723 ms 1,656 ms

首文本(SSE delta) 4,656 ms 6,996 ms 5,546 ms 4,933 ms

总耗时 7,069 ms 9,130 ms 7,968 ms 7,092 ms

输入 2 2 2 2

输出 421 715 659 677

缓存创建

4,262

5m: 0 / 1h: 4,262

4,753

5m: 0 / 1h: 4,753

3,002

5m: 0 / 1h: 3,002

4,762

5m: 0 / 1h: 4,762

缓存读取 0 0 1,772 0

输出速度 burst 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 17:42:13 ~ 17:42:18 (1.8s) 17:39:00 ~ 17:39:10 (9.3s) 17:42:15 ~ 17:42:20 (2.5s) 17:37:55 ~ 17:38:05 (8.4s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=806811083964357289403070336324463231351281799055664175317694396950715144

HTTP 头(TTFB) - 7,370 ms 1,217 ms 2,153 ms

首文本(SSE delta) - 7,399 ms 1,219 ms 7,149 ms

总耗时 - 8,346 ms 1,586 ms 7,333 ms

输入 - 211 211 211

输出 - 19 19 578

缓存创建

5m: 0 / 1h: 0

缓存读取 - 0 0 0

输出速度 - - - burst

响应

失败

评测失败

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

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

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

总计

总输入(含缓存)

129,701

仅前 6 步

132,447

132,388

132,372

总输出

504

仅前 6 步

819

783

1,354

总缓存创建

35,725

5m: 0 / 1h: 35,725

仅前 6 步

36,726

5m: 0 / 1h: 36,726

4,733

5m: 0 / 1h: 4,733

35,079

5m: 0 / 1h: 35,079

总缓存读取

92,540

仅前 6 步

94,049

125,982

95,620

输出速度

burst

仅前 6 步

48.9 tok/s

53.0 tok/s

63.2 tok/s

总执行时间

22.3s

仅前 6 步

1m 26.5s

28.7s

31.8s

总墙钟时间

15m 28.3s

含步骤间等待

仅前 6 步

12m 22.0s

含步骤间等待

15m 24.2s

含步骤间等待

11m 12.9s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2893

Anthropic: Claude Opus 4.8 (部分)

$0.3054

Anthropic: Claude Opus 4.8

$0.1205

Anthropic: Claude Opus 4.8

$0.3093

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 knowledge_recall digit_count digit_count digit_count

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

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

CDN - - - -

ID 格式 msg_01HNhWFvTRQbe9D3CxoznrFq msg_011nzYHFM7WQaivB4HJ2rpQ3 msg_01SVsFo3jMYMgZyPnKwDrcFe msg_01X6uo1XqSWij3GopBWfWVpo

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: a0d9523c7845dcfa-LAX server: Photon-Edge set-cookie: _cfuvid=xvtDQwTN6zAYovh.DLwAI5vpCG92H28byvuRBcAmDNw-1781775507.918988-1.0.1.1-8W.8UA7drPRzoEo.EWrET8PAmhbeNXQKGtlvicv.qsY; HttpOnly; SameSite=None; Secure; Path=/; Domain=api.anthropic.com	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	1781775507882-oe0efkxral0ad req_011CcAW4TNDT5t5aZGgAbDH7 a0d9523c7845dcfa-LAX
O-Max	server: nginx set-cookie: server_name_session=9a42ff809bdc7f49eaa4735d20f97eff; Max-Age=86400; httponly; path=/ x-new-api-version: v1.0.0-rc.11 x-oneapi-request-id: 202606180939012337316818268d9d6KzHTp742	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-idbody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—
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_011CcAWMJF1fsQh6CSbRt4iU
O-Max	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_011CcAW28HT7fDyTRX4AtmCa

协议指纹评分（机器自动）

总分 / 100 96 100 95

基线状态已对照基线已对照基线已对照基线

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

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

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

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

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

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

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

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

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

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

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

流式投递 — 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

综合结论

与基线相似度执行失败，不评分 96 v3.24.0 基准（参考） 95 v3.24.0

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 3177 ms 13425 ms 3753 ms 4214 ms

输出速度 burst 48.9 tok/s 53.0 tok/s 63.2 tok/s

完成步骤 5/6 成功 · 1 失败 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 ⓘ

$0.2893 (部分)

$0.3054

$0.1205

$0.3093

缓存读取占比 ⓘ 71% (部分) 71% 95% 72%

揭晓

通道

服务商 SSSAiCode ↗

通道O-Max

服务商 LinkAPI ↗

通道O-Max

官方基准

服务商 SAIAi ↗

通道O-Max

返回列表