模型对比 | RPDiag

测试说明点击展开 5 步 quick-probe-v1 详解

前 4 步同一会话（交互 TUI 多轮），第 5、6 步各自独立新会话，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 1-3 步的延迟在交互进程内发生，让缓存前缀在墙钟里老化）。第 3 步（cutoff）累计跨度 > 5 分钟保住 sliding cache 检测；总跨度 > 6 分钟。第 5 步（知识召回）与第 6 步（档位）都是独立新会话（第 6 步不复用上下文，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

2026/07/02 运行日期

claude-opus-4-8 官方基准

1 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

SSSAiCode / R-Kiro

不可测：未取得可评分响应（如 403 / 上游错误）

步骤 / 指标

ClaudeAPI

O-Bdrk

官方基准

SAIAi

O-Max

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 v2.1.195 claude-cli v2.1.195 claude-cli v2.1.195 claude-cli v2.1.195

偏差标红基于官方基准

步骤 1: ping text

时间 23:20:37 ~ 23:31:18 (638.0s) 23:20:45 ~ 23:28:51 (484.5s) 23:20:41 ~ 23:27:37 (415.2s) 23:20:39 ~ 23:31:00 (617.5s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 5,004 ms 1,975 ms 1,901 ms 4,391 ms

首文本(SSE delta) 5,005 ms 1,977 ms 1,902 ms 4,393 ms

总耗时 5,210 ms 2,057 ms 2,077 ms 4,481 ms

输入 2 2,579 2,579 2

输出 4 4 4 4

缓存创建

4,370

5m: 0 / 1h: 4,370

5m: 0 / 1h: 0

1,793

5m: 0 / 1h: 1,793

15,382

5m: 0 / 1h: 15,382

缓存读取 14,410 16,203 14,410 0

响应

pong

pong

pong

pong

步骤 2: identity text

时间 23:20:37 ~ 23:31:18 (638.0s) 23:20:45 ~ 23:28:51 (484.5s) 23:20:41 ~ 23:27:37 (415.2s) 23:20:39 ~ 23:31:00 (617.5s)

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。严格按以下 3 行格式输出，全部小写英文，不要任何其他文字、解释、标点：
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) 3,682 ms 3,387 ms 2,412 ms 4,198 ms

首文本(SSE delta) 4,267 ms 3,388 ms 2,413 ms 4,199 ms

总耗时 4,366 ms 3,756 ms 2,848 ms 5,846 ms

输入 2 2 2 2

输出 28 28 28 28

缓存创建

191

5m: 0 / 1h: 191

5m: 0 / 1h: 0

2,768

5m: 0 / 1h: 2,768

191

5m: 0 / 1h: 191

缓存读取 18,780 18,971 16,203 15,382

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 23:20:37 ~ 23:31:18 (638.0s) 23:20:45 ~ 23:28:51 (484.5s) 23:20:41 ~ 23:27:37 (415.2s) 23:20:39 ~ 23:31:00 (617.5s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 4,376 ms 2,024 ms 2,210 ms 5,321 ms

首文本(SSE delta) 4,378 ms 2,025 ms 2,211 ms 5,322 ms

总耗时 4,573 ms 2,150 ms 2,758 ms 5,507 ms

输入 2 2 2 2

输出 6 6 6 6

缓存创建

5m: 0 / 1h: 54

5m: 0 / 1h: 0

5m: 0 / 1h: 54

缓存读取 18,971 19,025 18,971 15,573

响应

2026-01

2026-01

2026-01

2026-01

步骤 4: identity_free text

时间 23:20:37 ~ 23:31:18 (638.0s) 23:20:45 ~ 23:28:51 (484.5s) 23:20:41 ~ 23:27:37 (415.2s) 23:20:39 ~ 23:31:00 (617.5s)

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) 5,888 ms 6,070 ms 2,672 ms 8,568 ms

首文本(SSE delta) 5,889 ms 6,077 ms 2,672 ms 8,569 ms

总耗时 7,263 ms 7,142 ms 3,491 ms 9,509 ms

输入 2 2 2 2

输出 60 68 66 58

缓存创建

4,654

5m: 0 / 1h: 4,654

5m: 0 / 1h: 31

5m: 0 / 1h: 39

缓存读取 14,410 19,025 19,025 15,627

输出速度 43.7 tok/s 63.8 tok/s 80.6 tok/s 61.7 tok/s

响应

ææ¯ Claude Codeï¼Anthropic å¼åçå®æ¹å½ä»¤è¡ç¼ç¨å©æï¼åºäº Claude Opus 4.8 æ¨¡åï¼å¸®å©ä½ å®æè½¯ä»¶å¼åä»»å¡ã

ææ¯ Claudeï¼ç± Anthropic å¼åç AI å©æãå½åæ¨¡åæ¯ Claude Opus 4.8

ææ¯ Claude Codeï¼ç± Anthropic å¼åçå®æ¹å½ä»¤è¡ AI ç¼ç¨å©æï¼åºäº Claude Opus 4.8 æ¨¡åï¼å¸®å©ä½ å®æè½¯ä»¶å·¥ç¨ä»»å¡ã

ææ¯ Claudeï¼ç± Anthropic å¼åçäººå·¥æºè½å©æï¼åºäº Claude Agent SDK æå»ºï¼å¯åå©å®æè½¯ä»¶å·¥ç¨çåç±»ä»»å¡ã

步骤 5: knowledge_recall text

时间 23:32:39 ~ 23:32:59 (19.4s) 23:30:10 ~ 23:30:30 (17.9s) 23:29:53 ~ 23:30:13 (18.3s) 23:34:24 ~ 23:34:49 (20.9s)

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) 2,261 ms 1,578 ms 1,507 ms 3,775 ms

首文本(SSE delta) 3,671 ms 2,589 ms 3,001 ms 7,722 ms

总耗时 8,672 ms 7,807 ms 8,394 ms 9,612 ms

输入 2 2 2 2

输出 685 699 708 698

缓存创建

3,153

5m: 0 / 1h: 3,153

5m: 0 / 1h: 0

3,153

5m: 0 / 1h: 3,153

33,973

5m: 0 / 1h: 33,973

缓存读取 0 3,153 0 0

输出速度 137.0 tok/s 134.0 tok/s 131.3 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

{"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 23:34:01 ~ 23:34:06 (4.2s) 23:33:01 ~ 23:33:06 (3.2s) 23:32:09 ~ 23:32:14 (2.8s) 23:36:59 ~ 23:37:05 (4.2s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=361249916168943309492568264183288542247291853016652038730063403804845121

HTTP 头(TTFB) 2,036 ms 1,801 ms 1,450 ms 2,609 ms

首文本(SSE delta) 2,037 ms 1,802 ms 1,451 ms 2,610 ms

总耗时 2,434 ms 2,100 ms 1,784 ms 2,904 ms

输入 211 211 211 211

输出 19 19 19 19

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 0 0

响应

{"n":72,"c":[8,3,7]}

{"n":72,"c":[6,3,7]}

{"n":72,"c":[6,4,7]}

{"n":72,"c":[7,3,7]}

总计

总输入(含缓存) 79,214 79,206 79,214 96,442

总输出 802 824 831 813

总缓存创建

12,422

5m: 0 / 1h: 12,422

5m: 0 / 1h: 31

7,807

5m: 0 / 1h: 7,807

49,639

5m: 0 / 1h: 49,639

总缓存读取 66,571 76,377 68,609 46,582

输出速度 116.9 tok/s 122.1 tok/s 124.6 tok/s 61.7 tok/s

总执行时间 42m 55.6s 32m 38.9s 28m 2.0s 41m 35.0s

总墙钟时间

13m 28.6s

含步骤间等待

12m 21.3s

含步骤间等待

11m 33.2s

含步骤间等待

16m 25.5s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.1321

Anthropic: Claude Opus 4.8

$0.0730

Anthropic: Claude Opus 4.8

$0.1179

Anthropic: Claude Opus 4.8

$0.3550

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 (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_01BKcr6ZhR9RXNV7fjEyRNib msg_018UWinoBTefzwcVQY9YPf44 msg_01DXAoTauVrGymiadZfjgd6d msg_014Vc2WiGpAzRyrKbTBcoDW1

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-Bdrk	x-new-api-version: v1.0.0-rc.11 x-oneapi-request-id: 202607021534029185059278268d9d6aDoGvWPm	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_011CcdURZ6TNZos4qJzaCzvp
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_011CcdUMib6wRjaVAmBAFzZa
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	202607021537014723463688268d9d6w80UbcQa

协议指纹评分（机器自动）

总分 / 100 97 100 100 88

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 10.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20 9.2 /10 ×20

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

延迟基线 7.8 /10 ×15 10.0 /10 ×15 10.0 /10 ×15 7.6 /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 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 10.0 /10 ×13

原生 msg-ID 10.0 /10 ×8 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 0.0 /10 ×8

身份(结构) 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 10.0 /10 ×6 10.0 /10 ×6 10.0 /10 ×6

知识截止 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

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

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

Req-ID 透传 0.0 /10 ×2 10.0 /10 ×2 10.0 /10 ×2 0.0 /10 ×2

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

流式投递未计入未计入未计入未计入

综合结论

与基线相似度 97 v3.27.0 基准（参考） 100 v3.27.0 88 v3.27.0

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 Thinking 体量 thinking_volume_match
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 5420 ms 4169 ms 3559 ms 6310 ms

输出速度 116.9 tok/s 122.1 tok/s 124.6 tok/s 61.7 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 ⓘ

$0.1321

$0.0730

$0.1179

$0.3550

缓存读取占比 ⓘ 84% 96% 87% 48%

揭晓

通道

服务商 ClaudeAPI ↗

通道O-Bdrk

官方基准

服务商 SAIAi ↗

通道O-Max

服务商 AnyAIGC ↗

通道M-

返回列表