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

2026/06/19 运行日期

claude-sonnet-4-6 官方基准

1 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

InfAI / R-Kiro
```
评测失败
```

步骤 / 指标

RunAPI

O-Max

FoxCode

R-ultra

Novaris

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 02:29:48 ~ 02:29:53 (4.1s) 02:29:50 ~ 02:29:55 (4.1s) 02:29:46 ~ 02:29:51 (3.4s) 02:29:54 ~ 02:29:59 (4.5s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,446 ms 2,484 ms 2,323 ms 3,295 ms

首文本(SSE delta) 2,449 ms 2,486 ms 2,324 ms 3,296 ms

总耗时 2,682 ms 3,072 ms 2,600 ms 3,410 ms

输入 3 74 3 3

输出 5 15 5 5

缓存创建

32,318

5m: 0 / 1h: 32,318

32,942

5m: 32,942 / 1h: 0

32,318

5m: 0 / 1h: 32,318

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 02:32:32 ~ 02:32:42 (5.1s) 02:33:29 ~ 02:33:39 (5.2s) 02:33:15 ~ 02:33:20 (3.4s) 02:33:22 ~ 02:33:27 (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。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) 3,167 ms 2,775 ms 2,192 ms 2,110 ms

首文本(SSE delta) 3,169 ms 2,941 ms 2,193 ms 2,112 ms

总耗时 3,847 ms 4,301 ms 2,520 ms 2,589 ms

输入 3 41 3 3

输出 22 71 22 22

缓存创建

178

5m: 0 / 1h: 178

33,478

5m: 33,478 / 1h: 0

178

5m: 0 / 1h: 178

178

5m: 0 / 1h: 178

缓存读取 32,318 0 32,318 32,318

输出速度 - 52.2 tok/s - -

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 02:35:10 ~ 02:35:15 (3.5s) 02:37:26 ~ 02:37:36 (5.3s) 02:35:12 ~ 02:35:17 (4.0s) 02:36:57 ~ 02:37:02 (4.6s)

提示词

查看本步 Prompt

你掌握的世界知识截止于何时？YYYY-MM 格式作答，不解释。

HTTP 头(TTFB) 2,096 ms 3,747 ms 2,895 ms 3,198 ms

首文本(SSE delta) 2,097 ms 3,748 ms 2,896 ms 3,354 ms

总耗时 2,407 ms 4,390 ms 3,193 ms 3,537 ms

输入 3 89 3 3

输出 8 49 8 8

缓存创建

5m: 0 / 1h: 55

33,848

5m: 33,848 / 1h: 0

5m: 0 / 1h: 38

缓存读取 32,496 0 32,496 32,496

输出速度 - 76.3 tok/s - -

响应

2025-08

2025-08

2025-08

2025-08

步骤 4: identity_free text

时间 02:36:34 ~ 02:36:39 (3.7s) 02:40:32 ~ 02:40:42 (5.2s) 02:38:31 ~ 02:38:42 (7.1s) 02:39:11 ~ 02:39:16 (4.1s)

提示词

查看本步 Prompt

用中文说说你的身份：什么产品？由谁打造？50 字以内。

HTTP 头(TTFB) 1,976 ms 3,100 ms 5,743 ms 2,259 ms

首文本(SSE delta) 1,977 ms 3,140 ms 5,744 ms 2,261 ms

总耗时 2,650 ms 4,381 ms 6,311 ms 3,077 ms

输入 3 42 3 3

输出 39 81 41 52

缓存创建

5m: 0 / 1h: 39

34,305

5m: 34,305 / 1h: 0

5m: 0 / 1h: 41

5m: 0 / 1h: 35

缓存读取 32,551 0 32,534 32,534

输出速度 57.9 tok/s 65.3 tok/s 72.3 tok/s 63.7 tok/s

响应

我是 Claude，由 Anthropic 打造的 AI 助手。当前版本为 Claude Sonnet 4.6。

我是 Kiro，由 Anthropic 开发的 AI 助手，基于 Claude 模型构建，专注于软件工程任务，同时也能协助写作、分析、规划等专业工作。

我叫 Claude，是由 Anthropic 开发的 AI 助手，当前版本为 Claude Sonnet 4.6。

我是Claude，由Anthropic开发的AI助手。我基于大型语言模型构建，旨在提供安全、有帮助的对话和任务协助。

步骤 5: knowledge_recall text

时间 02:38:00 ~ 02:38:05 (4.7s) 02:43:16 ~ 02:43:41 (21.5s) 02:42:11 ~ 02:42:21 (5.6s) 02:41:54 ~ 02:41:59 (4.4s)

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) 1,456 ms 2,130 ms 2,548 ms 1,641 ms

首文本(SSE delta) 1,457 ms 2,131 ms 2,549 ms 1,643 ms

总耗时 3,733 ms 20,636 ms 4,775 ms 3,484 ms

输入 3 90 3 3

输出 198 1,670 198 205

缓存创建

8,018

5m: 0 / 1h: 8,018

9,635

5m: 9,635 / 1h: 0

8,005

5m: 0 / 1h: 8,005

8,010

5m: 0 / 1h: 8,010

缓存读取 0 0 0 0

输出速度 87.0 tok/s 90.2 tok/s 88.9 tok/s 111.4 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 02:42:04 ~ 02:42:29 (22.0s) 02:46:22 ~ 02:46:57 (30.8s) 02:43:36 ~ 02:43:56 (15.8s) 02:45:20 ~ 02:45:35 (10.8s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=826456577045713744121485926434116594379650396184708567835485422334901490

HTTP 头(TTFB) 1,339 ms 5,616 ms 1,437 ms 1,440 ms

首文本(SSE delta) 1,340 ms 5,617 ms 1,445 ms 1,442 ms

总耗时 20,996 ms 30,012 ms 14,985 ms 9,826 ms

输入 164 34 164 164

输出 1,586 2,342 1,221 742

缓存创建

5m: 0 / 1h: 0

176

5m: 176 / 1h: 0

5m: 0 / 1h: 0

缓存读取 0 0 0 0

输出速度 80.7 tok/s 96.0 tok/s 90.2 tok/s 88.5 tok/s

响应

## Counting digits in s

**s** = `826456577045713744121485926434116594379650396184708567835485422334901490`

### Step-by-step indexing (groups of 10):
| Pos | Digits |
|-----|--------|
| 1–10 | 8 2 6

```
s = "826456577045713744121485926434116594379650396184708567835485422334901490"
```

Groups of 10:
- `8264565770` → 0:1, 7:2, 9:0
- `4571374412` → 0:0, 7:2, 9:0
- `1485926434` → 0:0, 7:0, 9:1
- `11

## Counting digits in s

**s** = `826456577045713744121485926434116594379650396184708567835485422334901490`

### Step-by-step parsing (groups of 10):

| Pos | Digits | 0s | 7s | 9s |
|-----|--------|-

```json
{"n":72,"c":[5,7,6]}
```

**Verification breakdown:**

| Digit | Positions (1-indexed) | Count |
|---|---|---|
| **0** | 10, 42, 50, 68, 72 | **5** |
| **7** | 8, 9, 13, 16, 38, 49, 54 | **7**

总计

总输入(含缓存) 138,152 144,754 138,107 138,106

总输出 1,858 4,228 1,495 1,034

总缓存创建

40,608

5m: 0 / 1h: 40,608

144,384

5m: 144,384 / 1h: 0

40,580

5m: 0 / 1h: 40,580

40,579

5m: 0 / 1h: 40,579

总缓存读取 97,365 0 97,348 97,348

输出速度 80.6 tok/s 91.3 tok/s 89.4 tok/s 90.5 tok/s

总执行时间 43.0s 1m 12.0s 39.2s 32.0s

总墙钟时间

12m 41.0s

含步骤间等待

17m 7.1s

含步骤间等待

14m 10.0s

含步骤间等待

15m 40.9s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2099

Anthropic: Claude Sonnet 4.6

$0.4029

Anthropic: Claude Sonnet 4.6

$0.2043

Anthropic: Claude Sonnet 4.6

$0.1974

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%) one-api (100%) one-api (100%) direct (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_01QnDwyiCRDC8ugRhtf7YDxi msg_01d6NvH4v3dJbGNGZAiJ29ch msg_01XeVCijQ7tgaiCnyqNWDwUG msg_017moiS6Kssq3mgXsMiPvebn

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	server: nginx/1.24.0 (Ubuntu) x-oneapi-request-id: 20260619024205385257872xOqKbLTh	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	—
R-ultra	cf-ray: a0dc74da4dfbe041-NRT server: cloudflare x-new-api-version: v0.0.0 x-oneapi-request-id: 202606181846231672802438268d9d6j7Ct2gfB	id_format: msg_prefix 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-idhdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsupstream:anthropic-direct-from-body	a0dc74da4dfbe041-NRT
O-Max	server: nginx x-oneapi-request-id: 202606181843371597300958268d9d66rib0o9s	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	—
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_011CcBDmH7VUeHe5LonqYuH8

协议指纹评分（机器自动）

总分 / 100 97 65 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 0.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20

缓存 TTL 一致 10.0 /10 ×15 0.0 /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 0.0 /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

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 0.0 /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 6.8 /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 65 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 6053 ms 11132 ms 5731 ms 4321 ms

输出速度 80.6 tok/s 91.3 tok/s 89.4 tok/s 90.5 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 ⓘ

$0.2099

$0.4029

$0.2043

$0.1974

缓存读取占比 ⓘ 70% 0% 70% 70%

揭晓

通道

服务商 RunAPI ↗

通道O-Max

服务商 FoxCode ↗

通道R-ultra

服务商 Novaris ↗

通道O-Max

官方基准

返回列表