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

2026/06/21 运行日期

claude-sonnet-4-6 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

ZeroCode / cc
```
评测失败
```

步骤 / 指标

OmniaKey

O-Max

FastCode

ClaudeCN

O-Max

官方基准

基本信息

服务名 Claude Code Claude Code Claude Code Claude Code

模型 claude-sonnet-4-6 claude-sonnet-4-6 claude-sonnet-4-6 claude-sonnet-4-6

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 01:00:17 ~ 01:00:27 (6.9s) 01:00:21 ~ 01:00:31 (6.0s) 01:00:15 ~ 01:00:25 (8.5s) 01:00:23 ~ 01:00:33 (8.0s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 4,093 ms 4,233 ms 4,902 ms 6,085 ms

首文本(SSE delta) 4,095 ms 4,234 ms 4,906 ms 6,087 ms

总耗时 4,438 ms 4,322 ms 5,966 ms 6,193 ms

输入 3 3 3 3

输出 5 5 5 5

缓存创建

32,318

5m: 0 / 1h: 32,318

32,318

5m: 0 / 1h: 32,318

32,319

5m: 0 / 1h: 32,319

32,318

5m: 0 / 1h: 32,318

缓存读取 0 0 0 0

响应

pong

pong

pong

pong

步骤 2: identity text

时间 01:01:33 ~ 01:01:43 (6.2s) 01:01:59 ~ 01:02:09 (9.1s) 01:04:08 ~ 01:04:23 (14.0s) 01:03:10 ~ 01:03:15 (3.6s)

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。自报身份，输出严格遵守以下 3 行格式（不允许 markdown、代码块、额外文本）：
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) 4,411 ms 5,358 ms 3,244 ms 2,029 ms

首文本(SSE delta) 4,997 ms 5,362 ms 3,245 ms 2,036 ms

总耗时 4,997 ms 6,715 ms 3,684 ms 2,654 ms

输入 3 3 3 3

输出 22 22 22 22

缓存创建

192

5m: 0 / 1h: 192

32,510

5m: 0 / 1h: 32,510

192

5m: 0 / 1h: 192

192

5m: 0 / 1h: 192

缓存读取 32,318 0 32,319 32,318

响应

vendor: anthropic
brand: claude
model: claude-sonnet-4-6

vendor: anthropic
brand: claude
model: claude-sonnet-4-6

vendor: anthropic
brand: claude
model: claude-sonnet-4-6

vendor: anthropic
brand: claude
model: claude-sonnet-4-6

步骤 3: cutoff text

时间 01:04:42 ~ 01:04:52 (7.3s) 01:04:10 ~ 01:04:20 (5.2s) 01:05:36 ~ 01:05:46 (5.6s) 01:04:53 ~ 01:04:58 (4.3s)

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) 6,067 ms 3,392 ms 3,642 ms 2,927 ms

首文本(SSE delta) 6,069 ms 3,393 ms 3,643 ms 3,207 ms

总耗时 6,069 ms 3,873 ms 4,542 ms 3,306 ms

输入 3 3 3 3

输出 8 8 8 8

缓存创建

5m: 0 / 1h: 57

5m: 0 / 1h: 38

缓存读取 32,510 32,510 32,511 32,510

响应

2025-08

2025-08

2025-08

2025-08

步骤 4: identity_free text

时间 01:07:48 ~ 01:07:58 (7.3s) 01:05:49 ~ 01:06:00 (6.6s) 01:08:13 ~ 01:08:24 (6.6s) 01:07:16 ~ 01:07:26 (4.9s)

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) 4,489 ms 3,823 ms 3,948 ms 2,845 ms

首文本(SSE delta) 4,490 ms 3,825 ms 3,949 ms 3,282 ms

总耗时 4,929 ms 5,479 ms 5,485 ms 3,899 ms

输入 3 3 3 3

输出 51 55 48 66

缓存创建

5m: 0 / 1h: 41

5m: 0 / 1h: 35

5m: 0 / 1h: 41

5m: 0 / 1h: 43

缓存读取 32,567 32,567 32,549 32,548

输出速度 116.2 tok/s 33.3 tok/s 31.3 tok/s 107.0 tok/s

响应

我是 Claude，由 Anthropic 开发的 AI 助手。当前版本为 Claude Sonnet 4.6，属于 Claude 4 系列产品。

我是 Claude，由 Anthropic 开发的 AI 助手。我基于大型语言模型构建，致力于提供安全、有益的对话和任务协助。

我叫 Claude，是由 Anthropic 公司开发的 AI 助手。当前使用的模型版本是 claude-sonnet-4-6。

我是 Claude，由 Anthropic 公司开发的 AI 助手。我擅长回答问题、分析代码、写作和推理，致力于提供安全、有帮助的人工智能服务。

步骤 5: knowledge_recall text

时间 01:09:21 ~ 01:09:31 (5.7s) 01:07:25 ~ 01:07:35 (6.1s) 01:12:19 ~ 01:12:29 (7.5s) 01:10:31 ~ 01:10:36 (4.8s)

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) 2,706 ms 2,284 ms 2,421 ms 1,802 ms

首文本(SSE delta) 2,708 ms 2,285 ms 2,423 ms 1,804 ms

总耗时 4,672 ms 4,642 ms 5,367 ms 3,675 ms

输入 3 3 3 3

输出 204 204 204 204

缓存创建

8,048

5m: 0 / 1h: 8,048

8,046

5m: 0 / 1h: 8,046

8,026

5m: 0 / 1h: 8,026

8,046

5m: 0 / 1h: 8,046

缓存读取 0 0 0 0

输出速度 103.9 tok/s 86.6 tok/s 69.3 tok/s 109.0 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 01:12:07 ~ 01:12:37 (28.5s) 01:09:39 ~ 01:09:59 (16.6s) 01:16:12 ~ 01:16:32 (17.1s) 01:11:54 ~ 01:12:14 (17.9s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=568972680554404782155239797002503334467670369103287914499655990785766384

HTTP 头(TTFB) 2,649 ms 1,252 ms 2,058 ms 1,217 ms

首文本(SSE delta) 2,650 ms 1,253 ms 3,859 ms 1,223 ms

总耗时 25,871 ms 15,039 ms 15,883 ms 16,810 ms

输入 164 164 164 164

输出 1,877 1,438 1,131 1,657

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 0 0

输出速度 80.8 tok/s 104.3 tok/s 94.1 tok/s 106.3 tok/s

响应

## Counting digits in s

**s** = `568972680554404782155239797002503334467670369103287914499655990785766384`

### Length
Grouping into 10s:
`5689726805` `5444047821` `5523979700` `2503334467` `67036910

## Counting digits in `s`

**String:** `568972680554404782155239797002503334467670369103287914499655990785766384`

### Length
Splitting into groups of 9:
`568972680` | `554404782` | `155239797` | `002

```json
{"n":72,"c":[8,9,9]}
```

**Verification:**

```
s = 568972680554404782155239797002503334467670369103287914499655990785766384
     1234567890123456789012345678901234567890123456789012345678901

## Counting digits in s

**s = 568972680554404782155239797002503334467670369103287914499655990785766384**

### Length (groups of 10):
```
5689726805  → 10
5440478215  → 10
5239797002  → 10
5033344676

总计

总输入(含缓存) 138,230 138,222 138,174 138,192

总输出 2,167 1,732 1,418 1,962

总缓存创建

40,656

5m: 0 / 1h: 40,656

72,966

5m: 0 / 1h: 72,966

40,616

5m: 0 / 1h: 40,616

40,637

5m: 0 / 1h: 40,637

总缓存读取 97,395 65,077 97,379 97,376

输出速度 83.2 tok/s 95.4 tok/s 83.8 tok/s 106.6 tok/s

总执行时间 1m 1.9s 49.5s 59.4s 43.6s

总墙钟时间

12m 20.0s

含步骤间等待

9m 38.1s

含步骤间等待

16m 17.1s

含步骤间等待

11m 51.1s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2147

Anthropic: Claude Sonnet 4.6

$0.3197

Anthropic: Claude Sonnet 4.6

$0.2033

Anthropic: Claude Sonnet 4.6

$0.2116

Anthropic: Claude Sonnet 4.6

通道指纹

来源步骤 digit_count digit_count digit_count digit_count

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

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

CDN cloudflare - cloudflare -

ID 格式 msg_01EQpmz6Ks9BKJNvBVrvSRt3 msg_bdrk_014ybY2svyxmFaEGhaqSz8jG msg_01H6MYd8Ti96ivbnrGit3zdW msg_018DUVrwrZQP3gcQ41MNrkpa

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: a0ec65969f94d3db-AMS server: cloudflare x-oneapi-request-id: 202606201712104996460808268d9d67NGDg0pX	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	a0ec65969f94d3db-AMS
cc	server: nginx	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: false has_inference_geo: false model_has_date_suffix: false usage_has_cache_fields: true	body:id:msg_prefixbody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—
O-Max	cf-ray: a0ec6b896d1a17dd-AMS server: cloudflare x-oneapi-request-id: 20260621011614228843548uiZzkBgY	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	a0ec6b896d1a17dd-AMS
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_011CcEtFtwV6aAueEosYwp8F

协议指纹评分（机器自动）

总分 / 100 97 90 97 100

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 5.6 /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 10.0 /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 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14

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

延迟基线 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 10.0 /10 ×5 10.0 /10 ×5

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

综合结论

与基线相似度 97 v3.24.0 90 v3.24.0 97 v3.24.0 基准（参考）

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 流式投递 buffer_dump_match
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 8496 ms 6678 ms 6821 ms 6090 ms

输出速度 83.2 tok/s 95.4 tok/s 83.8 tok/s 106.6 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 ⓘ

$0.2147

$0.3197

$0.2033

$0.2116

缓存读取占比 ⓘ 70% 47% 70% 70%

揭晓

通道

服务商 OmniaKey ↗

通道O-Max

服务商 FastCode ↗

通道cc

服务商 ClaudeCN ↗

通道O-Max

官方基准

返回列表