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-haiku-4-5-20251001 官方基准

2 通道分数 < 70 另含 1 任务失败

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

0-0 / O-Max
```
评测失败
```

步骤 / 指标

OmniaKey

O-Max

FoxCode

R-ultra

AnyAIGC

M-

官方基准

基本信息

服务名 Claude Code Claude Code Claude Code Claude Code

模型 claude-haiku-4-5-20251001 claude-haiku-4-5-20251001 claude-haiku-4-5-20251001 claude-haiku-4-5-20251001

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 03:22:13 ~ 03:22:23 (5.3s) 03:22:07 ~ 03:22:12 (3.7s) 03:22:09 ~ 03:22:19 (5.6s) 03:22:15 ~ 03:22:20 (3.1s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 3,608 ms 2,197 ms 3,817 ms 1,815 ms

首文本(SSE delta) 3,609 ms 2,198 ms 3,818 ms 1,817 ms

总耗时 4,091 ms 2,809 ms 4,280 ms 2,127 ms

输入 10 21,388 9,391 10

输出 42 50 2 52

缓存创建

6,166

5m: 0 / 1h: 6,166

5m: 0 / 1h: 0

缓存读取 15,379 0 0 21,545

输出速度 87.1 tok/s 81.8 tok/s - 167.7 tok/s

响应

pong

pong

pong

pong

步骤 2: identity text

时间 03:24:06 ~ 03:24:16 (6.2s) 03:24:48 ~ 03:24:58 (5.3s) 03:25:04 ~ 03:25:14 (5.0s) 03:23:47 ~ 03:23:52 (4.4s)

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) 3,259 ms 1,713 ms 3,819 ms 1,460 ms

首文本(SSE delta) 3,260 ms 1,714 ms 3,820 ms 1,462 ms

总耗时 5,120 ms 4,571 ms 3,820 ms 3,410 ms

输入 10 21,956 9,590 10

输出 190 356 25 209

缓存创建

236

5m: 0 / 1h: 236

5m: 0 / 1h: 0

246

5m: 0 / 1h: 246

缓存读取 21,545 0 0 21,545

输出速度 102.2 tok/s 124.6 tok/s - 107.3 tok/s

响应

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

步骤 3: cutoff text

时间 03:28:05 ~ 03:28:16 (6.5s) 03:26:44 ~ 03:26:49 (3.9s) 03:26:46 ~ 03:26:51 (3.7s) 03:27:50 ~ 03:27:55 (3.4s)

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) 4,947 ms 2,372 ms 2,546 ms 1,958 ms

首文本(SSE delta) 4,948 ms 2,374 ms 2,547 ms 1,959 ms

总耗时 5,390 ms 3,121 ms 2,570 ms 2,429 ms

输入 10 22,676 9,631 10

输出 76 87 5 72

缓存创建

232

5m: 0 / 1h: 232

5m: 0 / 1h: 0

249

5m: 0 / 1h: 249

缓存读取 21,781 0 0 21,791

输出速度 171.9 tok/s 116.5 tok/s - 153.2 tok/s

响应

2025-02

2025-02

2025-05

2025-02

步骤 4: identity_free text

时间 03:30:18 ~ 03:30:28 (6.0s) 03:29:55 ~ 03:30:05 (5.1s) 03:30:20 ~ 03:30:30 (5.1s) 03:31:35 ~ 03:31:45 (8.7s)

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) 2,759 ms 2,000 ms 3,853 ms 1,417 ms

首文本(SSE delta) 2,760 ms 2,001 ms 3,854 ms 1,419 ms

总耗时 4,871 ms 4,226 ms 3,862 ms 7,676 ms

输入 10 23,110 9,662 10

输出 219 474 25 868

缓存创建

116

5m: 0 / 1h: 116

5m: 0 / 1h: 0

114

5m: 0 / 1h: 114

缓存读取 22,013 0 0 22,040

输出速度 103.7 tok/s 213.0 tok/s - 138.7 tok/s

响应

我是Claude Haiku 4.5，属于Claude AI助手产品，由Anthropic公司开发。

我是Claude，由Anthropic公司开发的AI助手。我是一个大型语言模型，可以帮助用户进行编程、写作、分析、研究等各种专业工作。

我是由Anthropic开发的AI助手，名为Claude，旨在帮助用户解决技术和信息相关的问题。

我是Claude Haiku，由Anthropic开发的AI编程助手。可协助代码编写、调试、文件管理等软件工程任务。

步骤 5: knowledge_recall text

时间 03:31:51 ~ 03:32:16 (21.5s) 03:32:05 ~ 03:32:25 (19.3s) 03:34:22 ~ 03:34:27 (4.9s) 03:33:13 ~ 03:33:43 (25.4s)

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) 2,304 ms 1,842 ms 2,624 ms 1,481 ms

首文本(SSE delta) 2,305 ms 2,010 ms 2,625 ms 1,483 ms

总耗时 20,468 ms 18,510 ms 3,736 ms 24,260 ms

输入 10 11,300 10,290 10

输出 2,167 3,003 154 3,806

缓存创建

9,299

5m: 0 / 1h: 9,299

5m: 0 / 1h: 0

9,973

5m: 0 / 1h: 9,973

缓存读取 0 0 0 0

输出速度 119.3 tok/s 182.0 tok/s 138.6 tok/s 167.1 tok/s

响应

```json
{"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

```json
{"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

{"facts":{"super_bowl_lviii":"UNKNOWN","oscars2024_best_picture":"UNKNOWN","ucl_2024":"UNKNOWN","nba_2024":"UNKNOWN","copa_america_2024":"UNKNOWN","euro2024":"UNKNOWN","nobel_peace_2024":"UNKNOWN","wo

```json
{"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

步骤 6: digit_count text

时间 03:34:48 ~ 03:35:23 (34.8s) 03:33:40 ~ 03:33:55 (11.4s) 03:35:42 ~ 03:35:47 (4.6s) 03:35:00 ~ 03:35:15 (13.1s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=856053857701443385202962926473351567496191035564818331551554641813715071

HTTP 头(TTFB) 1,742 ms 1,158 ms 2,523 ms 792 ms

首文本(SSE delta) 1,743 ms 1,159 ms 2,524 ms 794 ms

总耗时 33,696 ms 10,765 ms 3,340 ms 12,123 ms

输入 193 209 165 193

输出 4,030 2,245 14 1,628

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 0 0

输出速度 126.1 tok/s 233.7 tok/s - 143.7 tok/s

响应

```json
{"n": 72, "c": [5, 6, 4]}
```

```json
{"n":72,"c":[5,6,4]}
```

{"n":87,"c":[2,1,4]}

```json
{"n":72,"c":[5,6,4]}
```

总计

总输入(含缓存) 97,010 100,639 48,729 97,746

总输出 6,724 6,215 225 6,635

总缓存创建

16,049

5m: 0 / 1h: 16,049

5m: 0 / 1h: 0

10,582

5m: 0 / 1h: 10,582

总缓存读取 80,718 0 0 86,921

输出速度 122.2 tok/s 191.0 tok/s 138.6 tok/s 154.0 tok/s

总执行时间 1m 20.2s 48.7s 28.9s 58.0s

总墙钟时间

13m 10.1s

含步骤间等待

11m 48.1s

含步骤间等待

13m 37.9s

含步骤间等待

12m 60.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.0620

Anthropic: Claude Haiku 4.5

$0.1317

Anthropic: Claude Haiku 4.5

$0.0499

Anthropic: Claude Haiku 4.5

$0.0553

Anthropic: Claude Haiku 4.5

通道指纹

来源步骤 digit_count digit_count digit_count digit_count

平台 one-api (100%) one-api (100%) direct (50%) direct (100%)

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

CDN cloudflare cloudflare - -

ID 格式 msg_01YbRz5eWcPiV3JXDHEmTwRJ msg_01WGCV4T4fjvSwDN7fk2qWUu 自定义 msg_01RXDYxQroNwqaiNM9dhyT2r

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: a13f9c50cf415925-AMS server: cloudflare x-oneapi-request-id: 202606301934502280351008268d9d6UxKH6dWB	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: true has_inference_geo: true model_has_date_suffix: true usage_has_cache_fields: true	hdr:x-oneapi-request-idhdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:model:date_suffixbody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a13f9c50cf415925-AMS
R-ultra	cf-ray: a13f9aa53f09e393-NRT server: cloudflare x-new-api-version: v0.0.0 x-oneapi-request-id: 202606301933414888058938268d9d613zZSjx4	id_format: msg_prefix has_service_tier: true has_inference_geo: true model_has_date_suffix: true usage_has_cache_fields: true	hdr:x-new-api-versionhdr:x-oneapi-request-idhdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:model:date_suffixbody:usage.cache_fieldsupstream:anthropic-direct-from-body	a13f9aa53f09e393-NRT
M-	server: nginx	id_format: non_standard has_service_tier: false has_inference_geo: false model_has_date_suffix: true usage_has_cache_fields: true	body:id:non_standardbody:model:date_suffixbody:usage.cache_fields	—
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: true usage_has_cache_fields: true	body:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:model:date_suffixbody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	req_011Cca1FxAqv7Lm8uuGDgddX

协议指纹评分（机器自动）

总分 / 100 97 69 35 100

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 0.0 /10 ×20 0.0 /10 ×20 10.0 /10 ×20

缓存 TTL 一致 10.0 /10 ×15 0.0 /10 ×15 0.0 /10 ×15 10.0 /10 ×15

延迟基线 7.3 /10 ×15 10.0 /10 ×15 7.1 /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 5.0 /10 ×14 5.0 /10 ×14 10.0 /10 ×14

sliding 5m 10.0 /10 ×13 5.0 /10 ×13 5.0 /10 ×13 10.0 /10 ×13

原生 msg-ID 10.0 /10 ×8 10.0 /10 ×8 0.0 /10 ×8 10.0 /10 ×8

系统提示纯净 10.0 /10 ×8 10.0 /10 ×8 0.0 /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 0.0 /10 ×6 10.0 /10 ×6

知识截止 10.0 /10 ×5 10.0 /10 ×5 5.0 /10 ×5 10.0 /10 ×5

inference_geo 10.0 /10 ×5 10.0 /10 ×5 0.0 /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 0.0 /10 ×2 0.0 /10 ×2 10.0 /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 v3.26.0 69 v3.26.0 35 v3.26.0 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 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 12273 ms 7334 ms 3601 ms 8671 ms

输出速度 122.2 tok/s 191.0 tok/s 138.6 tok/s 154.0 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 ⓘ

$0.0620

$0.1317

$0.0499

$0.0553

缓存读取占比 ⓘ 83% 0% 0% 89%

揭晓

通道

服务商 OmniaKey ↗

通道O-Max

服务商 FoxCode ↗

通道R-ultra

服务商 AnyAIGC ↗

通道M-

官方基准

返回列表