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

2026/07/01 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

TopRouterCN / O-Max
```
评测失败
```

步骤 / 指标

官方基准

LingxiCode

M-

LinkAPI

O-Max

ClaudeCN

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 v2.1.195 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 12:59:04 ~ 12:59:09 (4.2s) 12:58:56 ~ 12:59:06 (5.0s) 12:59:00 ~ 12:59:05 (3.3s) 12:58:58 ~ 12:59:08 (5.9s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,954 ms 2,676 ms 1,871 ms 3,062 ms

首文本(SSE delta) 2,956 ms 2,683 ms 1,913 ms 3,063 ms

总耗时 3,155 ms 3,284 ms 2,022 ms 3,465 ms

输入 2,326 2,372 2,326 2,326

输出 4 4 4 4

缓存创建

1,711

5m: 0 / 1h: 1,711

1,745

5m: 0 / 1h: 1,711

14,443

5m: 0 / 1h: 14,443

1,714

5m: 0 / 1h: 1,714

缓存读取 12,732 12,986 0 12,732

响应

pong

pong

pong

pong

步骤 2: identity text

时间 13:01:54 ~ 13:01:59 (3.1s) 13:01:42 ~ 13:01:47 (4.0s) 13:00:52 ~ 13:00:57 (3.2s) 13:01:05 ~ 13:01:15 (5.3s)

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) 1,652 ms 2,408 ms 1,806 ms 3,861 ms

首文本(SSE delta) 1,653 ms 2,411 ms 1,807 ms 4,102 ms

总耗时 2,042 ms 3,154 ms 2,234 ms 4,291 ms

输入 2 2 2 2

输出 28 28 28 28

缓存创建

2,502

5m: 0 / 1h: 2,502

4,300

5m: 0 / 1h: 4,216

2,502

5m: 0 / 1h: 2,502

4,215

5m: 0 / 1h: 4,215

缓存读取 14,443 12,986 14,443 12,732

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 13:05:56 ~ 13:06:06 (5.3s) 13:04:18 ~ 13:04:23 (3.7s) 13:03:35 ~ 13:03:40 (3.5s) 13:04:42 ~ 13:04:47 (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) 3,647 ms 2,502 ms 2,238 ms 2,859 ms

首文本(SSE delta) 3,680 ms 2,503 ms 2,239 ms 3,217 ms

总耗时 4,153 ms 2,795 ms 2,443 ms 3,331 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

缓存读取 16,945 17,286 16,945 16,947

响应

2026-01

2026-01

2026-01

2026-01

步骤 4: identity_free text

时间 13:09:08 ~ 13:09:13 (3.7s) 13:06:40 ~ 13:06:46 (4.6s) 13:07:00 ~ 13:07:05 (3.4s) 13:07:30 ~ 13:07:40 (5.6s)

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) 1,799 ms 2,764 ms 1,753 ms 2,766 ms

首文本(SSE delta) 1,800 ms 2,765 ms 1,754 ms 2,767 ms

总耗时 2,702 ms 3,844 ms 2,438 ms 4,274 ms

输入 2 2 2 2

输出 59 61 39 68

缓存创建

5m: 0 / 1h: 37

5m: 0 / 1h: 39

5m: 0 / 1h: 37

5m: 0 / 1h: 39

缓存读取 17,009 17,350 17,007 17,009

输出速度 65.4 tok/s 56.5 tok/s 57.0 tok/s 45.1 tok/s

响应

我是 Claude，属于 Claude 产品系列，由 Anthropic 公司开发。具体模型为 Opus 4.8（claude-opus-4-8）。

我是 Claude，由 Anthropic 开发的 AI 助手，基于 Claude Opus 4.8 模型，可协助编程、写作、分析等各类任务。

我是 Claude Opus 4.8，属于 Claude 产品系列，由 Anthropic 开发。

我是 Claude，由 Anthropic 开发的 AI 助手。当前基于 Opus 4.8 模型，可协助编程、写作、分析等各类任务，注重安全与实用。

步骤 5: knowledge_recall text

时间 13:13:08 ~ 13:13:23 (11.9s) 13:09:37 ~ 13:09:47 (8.9s) 13:08:24 ~ 13:08:39 (10.1s) 13:08:51 ~ 13:09:06 (11.3s)

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) 4,187 ms 2,051 ms 2,742 ms 3,158 ms

首文本(SSE delta) 5,610 ms 3,459 ms 4,606 ms 4,578 ms

总耗时 10,943 ms 7,950 ms 9,163 ms 10,232 ms

输入 2 2 2 2

输出 722 684 716 784

缓存创建

5,742

5m: 0 / 1h: 5,742

5,861

5m: 0 / 1h: 5,747

5,720

5m: 0 / 1h: 5,720

5,753

5m: 0 / 1h: 5,753

缓存读取 0 0 0 0

输出速度 135.4 tok/s burst burst 138.7 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 13:15:14 ~ 13:15:19 (4.5s) 13:13:02 ~ 13:13:07 (3.3s) 13:09:54 ~ 13:10:09 (13.6s) 13:12:36 ~ 13:12:51 (11.9s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=220732282848906260013137086422480128283363837783819314783040407118704910

HTTP 头(TTFB) 3,131 ms 2,076 ms 3,530 ms 9,045 ms

首文本(SSE delta) 3,133 ms 2,077 ms 4,414 ms 9,046 ms

总耗时 3,475 ms 2,442 ms 12,685 ms 9,740 ms

输入 211 215 2 211

输出 19 19 872 19

缓存创建

5m: 0 / 1h: 0

2,104

5m: 2,104 / 1h: 0

5m: 0 / 1h: 0

缓存读取 0 0 0 0

输出速度 - - 105.4 tok/s -

响应

{"n":70,"c":[8,5,2]}

{"n":70,"c":[7,6,3]}

{"n":70,"c":[11,7,3]}

{"n":70,"c":[8,6,3]}

总计

总输入(含缓存) 73,730 75,211 75,599 73,748

总输出 838 802 1,665 909

总缓存创建

10,056

5m: 0 / 1h: 10,056

12,008

5m: 0 / 1h: 11,775

24,868

5m: 2,104 / 1h: 22,764

11,783

5m: 0 / 1h: 11,783

总缓存读取 61,129 60,608 48,395 59,420

输出速度 125.3 tok/s 56.5 tok/s 101.7 tok/s 119.0 tok/s

总执行时间 32.6s 29.3s 37.2s 44.4s

总墙钟时间

16m 15.0s

含步骤间等待

14m 10.9s

含步骤间等待

11m 9.0s

含步骤间等待

13m 53.1s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.1271

Anthropic: Claude Opus 4.8

$0.1369

Anthropic: Claude Opus 4.8

$0.2280

Anthropic: Claude Opus 4.8

$0.1388

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 digit_count digit_count digit_count digit_count

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

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

CDN - cloudflare - cloudflare

ID 格式 msg_01RTzP9EfuQoYatSdR61zTgM msg_015VAbEvPKmFchVAFTK6QFor msg_013DJCyjcoXmf42SomJvS8W9 msg_0196KNJyaFVvdn5zc6UprzU2

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
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_011CcamWEhXfhqmhxvQ4E541
M-	cf-ray: a142eb52bb00e3bf-NRT server: cloudflare set-cookie: server_name_session=27b20898e77738b0712c6970171c2581; Max-Age=86400; httponly; path=/ x-oneapi-request-id: 20260701051303769253969dnIBr771	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	a142eb52bb00e3bf-NRT
O-Max	server: nginx set-cookie: server_name_session=c576e09cb6f84c78844c48bf7e3b48cb; Max-Age=86400; httponly; path=/ x-new-api-version: v1.0.0-rc.15 x-oneapi-request-id: 202607010509553309647938268d9d6YJvgEMcT	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	—
O-Max	cf-ray: a142eab9e80097c4-AMS server: cloudflare x-oneapi-request-id: 202607011312392518285347W7rQR2C	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	a142eab9e80097c4-AMS

协议指纹评分（机器自动）

总分 / 100 100 97 88 99

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 9.0 /10 ×20 10.0 /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 10.0 /10 ×15 10.0 /10 ×15 10.0 /10 ×15 9.8 /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 5.0 /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 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

流式投递 10.0 /10 ×5 10.0 /10 ×5 0.6 /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 10.0 /10 ×2 0.0 /10 ×2 0.0 /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

综合结论

与基线相似度基准（参考） 97 v3.26.0 88 v3.26.0 99 v3.26.0

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 Thinking 体量 thinking_volume_match
14 缓存连续 cache_continuity_intra
15 知识召回 knowledge_recall_match
16 延迟基线 latency_baseline_match
17 原生 msg-ID anthropic_msg_id_format
18 Sliding 缓存 cache_sliding_correctness
19 结构化身份 identity_structured_match
20 信封自报一致 envelope_self_report_match
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 4412 ms 3912 ms 5164 ms 5889 ms

输出速度 125.3 tok/s 56.5 tok/s 101.7 tok/s 119.0 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 ⓘ

$0.1271

$0.1369

$0.2280

$0.1388

缓存读取占比 ⓘ 83% 81% 64% 81%

揭晓

通道

官方基准

服务商 LingxiCode ↗

通道M-

服务商 LinkAPI ↗

通道O-Max

服务商 ClaudeCN ↗

通道O-Max

返回列表