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

2026/06/30 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

FastCode / cc
```
评测失败
```

步骤 / 指标

OmniaKey

O-Max

SSSAiCode

O-Max-L2

官方基准

ByteCatCode

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 12:01:40 ~ 12:01:50 (5.6s) 12:01:36 ~ 12:01:41 (3.4s) 12:01:42 ~ 12:01:47 (3.1s) 12:01:34 ~ 12:01:44 (6.3s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,880 ms 2,019 ms 1,838 ms 4,712 ms

首文本(SSE delta) 2,881 ms 2,154 ms 1,840 ms -

总耗时 3,355 ms 2,279 ms 2,016 ms 5,072 ms

输入 1,454 1,428 1,454 1,454

输出 4 4 4 4

缓存创建

30,222

5m: 0 / 1h: 30,222

29,761

5m: 0 / 1h: 29,761

30,222

5m: 0 / 1h: 30,222

30,380

5m: 0 / 1h: 30,380

缓存读取 0 0 0 0

响应

pong

pong

pong

pong

步骤 2: identity text

时间 12:05:32 ~ 12:05:37 (4.8s) 12:04:49 ~ 12:04:54 (3.5s) 12:03:40 ~ 12:03:50 (5.2s) 12:05:09 ~ 12:05:14 (4.2s)

提示词

查看本步 Prompt

以下是对你自身的 metadata 探针。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。按以下 3 行回答自己的身份，每行一个 label，不要寒暄不要解释：
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) 3,253 ms 2,046 ms 3,733 ms 2,475 ms

首文本(SSE delta) 3,254 ms 2,408 ms 3,735 ms -

总耗时 3,683 ms 2,549 ms 4,201 ms 3,103 ms

输入 2 2 2 2

输出 28 32 28 28

缓存创建

1,656

5m: 0 / 1h: 1,656

1,630

5m: 0 / 1h: 1,630

1,656

5m: 0 / 1h: 1,656

1,656

5m: 0 / 1h: 1,656

缓存读取 30,222 29,761 30,222 30,380

响应

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

vendor: anthropic
brand: claude
model: claude-opus-4-5-20251101

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 12:08:09 ~ 12:08:19 (5.0s) 12:07:11 ~ 12:07:16 (3.1s) 12:06:19 ~ 12:06:24 (4.3s) 12:08:27 ~ 12:08:32 (3.7s)

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) 3,791 ms 2,017 ms 3,030 ms 2,178 ms

首文本(SSE delta) 3,871 ms 2,191 ms 3,031 ms -

总耗时 3,872 ms 2,254 ms 3,303 ms 2,526 ms

输入 2 2 2 2

输出 6 6 6 6

缓存创建

5m: 0 / 1h: 64

5m: 0 / 1h: 58

5m: 0 / 1h: 62

5m: 0 / 1h: 64

缓存读取 31,878 31,391 31,878 32,036

响应

2026-01

2025-03

2026-01

2026-01

步骤 4: identity_free text

时间 12:12:00 ~ 12:12:10 (7.2s) 12:09:25 ~ 12:09:30 (4.7s) 12:10:20 ~ 12:10:25 (4.4s) 12:11:21 ~ 12:11:26 (4.5s)

提示词

查看本步 Prompt

用中文回答：你是谁？由谁开发？50 字以内。

HTTP 头(TTFB) 4,705 ms 2,208 ms 2,681 ms 2,494 ms

首文本(SSE delta) 4,706 ms 3,305 ms 2,683 ms -

总耗时 5,931 ms 3,445 ms 3,384 ms 3,444 ms

输入 2 2 2 2

输出 69 77 46 49

缓存创建

5m: 0 / 1h: 31

5m: 0 / 1h: 39

5m: 0 / 1h: 31

5m: 0 / 1h: 39

缓存读取 31,942 31,449 31,940 32,100

输出速度 56.3 tok/s - 65.6 tok/s 51.6 tok/s

响应

我是 Claude，一个由 Anthropic 开发的人工智能助手。我基于 Claude Opus 4.8 模型，可以帮助你完成编程、写作、分析等各类任务。

我是 Claude，由 Anthropic 开发的 AI 助手。我可以帮你处理软件工程任务、回答问题、分析代码等。这里运行在命令行环境中，专注协助你完成编程工作。

我是 Claude,一个由 Anthropic 开发的 AI 助手。当前模型是 Claude Opus 4.8。

我是 Claude，由 Anthropic 开发的 AI 助手，可协助编程、写作、分析与解答问题等多种任务。

步骤 5: knowledge_recall text

时间 12:14:50 ~ 12:15:05 (10.1s) 12:12:06 ~ 12:12:16 (5.5s) 12:14:12 ~ 12:14:22 (9.5s) 12:15:09 ~ 12:15:19 (9.6s)

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) 3,075 ms 1,670 ms 1,834 ms 2,375 ms

首文本(SSE delta) 6,396 ms 2,173 ms 5,579 ms -

总耗时 8,954 ms 4,547 ms 8,337 ms 8,308 ms

输入 2 2 2 2

输出 693 347 688 693

缓存创建

4,796

5m: 0 / 1h: 4,796

4,319

5m: 0 / 1h: 4,319

4,771

5m: 0 / 1h: 4,771

4,784

5m: 0 / 1h: 4,784

缓存读取 0 0 0 0

输出速度 burst 146.2 tok/s burst 116.8 tok/s

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 12:17:46 ~ 12:17:51 (3.6s) 12:13:50 ~ 12:13:55 (1.1s) 12:15:57 ~ 12:16:02 (3.1s) 12:16:31 ~ 12:16:36 (3.1s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=704591931120409950270022065460354898360817403448735508901385414537728704

HTTP 头(TTFB) 2,027 ms - 1,762 ms 1,325 ms

首文本(SSE delta) 2,028 ms - 1,765 ms -

总耗时 2,357 ms - 2,088 ms 1,748 ms

输入 211 - 211 211

输出 19 - 19 19

缓存创建

5m: 0 / 1h: 0

缓存读取 0 - 0 0

响应

{"n":72,"c":[12,5,8]}

失败

评测失败

{"n":72,"c":[12,3,5]}

{"n":72,"c":[9,4,4]}

总计

总输入(含缓存)

132,484

129,844

仅前 6 步

132,455

133,112

总输出

819

466

仅前 6 步

791

799

总缓存创建

36,769

5m: 0 / 1h: 36,769

35,807

5m: 0 / 1h: 35,807

仅前 6 步

36,742

5m: 0 / 1h: 36,742

36,923

5m: 0 / 1h: 36,923

总缓存读取

94,042

92,601

仅前 6 步

94,040

94,516

输出速度

56.3 tok/s

146.2 tok/s

仅前 6 步

65.6 tok/s

107.8 tok/s

总执行时间

36.2s

21.2s

仅前 6 步

29.5s

31.4s

总墙钟时间

16m 11.0s

含步骤间等待

12m 19.0s

含步骤间等待

仅前 6 步

14m 20.0s

含步骤间等待

15m 2.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.3057

Anthropic: Claude Opus 4.8

$0.2889

Anthropic: Claude Opus 4.8 (部分)

$0.3048

Anthropic: Claude Opus 4.8

$0.3064

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 digit_count knowledge_recall digit_count digit_count

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

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

CDN cloudflare - - -

ID 格式 msg_017xn7bxXBj1pSC8afiVgTmm msg_01F7XQnLL9joTHD6Pg8W8ivi msg_01CdRExg2KSP2pr9kU7yUnTh msg_01S1K4M2kC2CCPdWW945F4s4

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: a13a5d03ffa59887-AMS server: cloudflare x-oneapi-request-id: 202606300417486328958188268d9d6xvxyR2Mm	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-idhdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a13a5d03ffa59887-AMS
O-Max-L2	cf-ray: a13a54b60a0d6e62-LAX server: Photon-Edge set-cookie: _cfuvid=RD9NEAQnOhGVD8pJVjPXiUOMkrcUZE8TikPkDO93ng4-1782792728.0082533-1.0.1.1-inENW4H6RCqcX7xD.2wN4uDRAaaN8t90kzjYATpYsTc; 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	1782792727874-1okuamm3idhpt req_011CcYnt86svf5kzCoYCiatj a13a54b60a0d6e62-LAX
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_011CcYoBAtnGh8wDc6HYkN7z
O-Max	server: nginx/1.29.8 x-oneapi-request-id: 202606300416332329359398268d9d61OGVl5Zm	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	—

协议指纹评分（机器自动）

总分 / 100 98 100 98

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 — 10.0 /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 9.6 /10 ×5 — 10.0 /10 ×5 10.0 /10 ×5

流式投递 10.0 /10 ×5 — 10.0 /10 ×5 未计入

Req-ID 透传 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

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 98 v3.25.0 执行失败，不评分基准（参考） 98 v3.25.0

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 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 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 4692 ms 3015 ms 3888 ms 4034 ms

输出速度 56.3 tok/s 146.2 tok/s 65.6 tok/s 107.8 tok/s

完成步骤 6/6 成功 5/6 成功 · 1 失败 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 ⓘ

$0.3057

$0.2889 (部分)

$0.3048

$0.3064

缓存读取占比 ⓘ 71% 71% (部分) 71% 71%

揭晓

通道

服务商 OmniaKey ↗

通道O-Max

服务商 SSSAiCode ↗

通道O-Max-L2

官方基准

服务商 ByteCatCode ↗

通道O-Max

返回列表