docs: organize benchmarks and add Web Fetch results

[lanyue-llk]

内容

统一整理六种语言 README（英文、中文、日文、德文、法文、西班牙文）的评测结构：内容提取、自动化兼容性、资源开销与功能覆盖。

• Web Fetch 只展示五款浏览器，名称统一为 Moli，实测提交为 721bd2e。用结果表和简短摘要说明其综合相似度、代码保真度及 545/545 正文生成覆盖。
• 保留 browser-eval 复现链接；详细计分方法、实测版本与历史 Moli 结果留在报告中。
• 保留 Lexbench 图表、数据与出处；移除旧的 192-URL 抓取样例，单组 Agent 样例保留为折叠补充。

Stack

GitHub 原生依赖链：#1334 → #1335 → #1336 → #1337 → #1380 → #1428 → #1431 → #1432。

本 PR 直接依赖 #1337，GitHub base 为 codex/content-rendered-state。差异仅含英文 README.md 和 docs/ 下的五种语言 README。按从左到右的顺序合入；#1428 直接以本分支为 base；#1431 与 #1432 再逐层堆叠，每层只显示自己的产品修复。#919 不在这条依赖链中。

验证

• 六种语言五行结果的 30 项分数、正文数量和唯一最高值标记均与公开报告一致。
• GitHub Markdown 实际渲染通过；原有图片路径、折叠内容与 diff 检查通过。
• 仅文档修改，没有重新采集或改变评测结果。

[github-actions]

CI Regression Report

Source CI run · source state at render: in_progress · artifacts: 6/6

Check	Status	Signal
Release regression	✅	HEAD/base failures 0/0; raw binary 0.00%
Frontend differential	✅	1,020/1,020 cases matched; 0 issues
Agent episodes	✅	8/8 Moli episodes passed; 0 failures
Runtime/CDP contracts	✅	26 contract cases; 0 failures
CDP smoke	✅	52/52 groups passed; 621 scenarios
WebMainBench	✅	545/545 completed; 0 unexpected failures

Release regression — ✅ HEAD/base failures 0/0; raw binary 0.00%

Package and image size

Metric	Base	HEAD	Delta	Delta %
Raw binary	149.03 MiB	149.03 MiB	0 B	0.00%
Stripped binary	102.66 MiB	102.66 MiB	0 B	0.00%
gzip binary	51.15 MiB	51.15 MiB	−29 B	−0.000054%
Rootfs	155.94 MiB	155.94 MiB	0 B	0.00%
gzip rootfs	60.05 MiB	60.05 MiB	−4 B	−0.000006%

Startup latency and PSS

Case	Base p50	HEAD p50	Delta	Base PSS	HEAD PSS	Delta
Serve ready	27.42 ms	16.66 ms	−10.76 ms	5.23 MiB	6.81 MiB	+1.57 MiB
CDP first page	52.11 ms	50.40 ms	−1.71 ms	5.30 MiB	6.68 MiB	+1.38 MiB
CDP warm page	12.82 ms	17.11 ms	+4.29 ms	50.17 MiB	49.32 MiB	−879.50 KiB
CLI about:blank	15.08 ms	18.62 ms	+3.54 ms	4.44 MiB	3.92 MiB	−526.00 KiB
CLI local JS	25.40 ms	21.00 ms	−4.40 ms	8.41 MiB	3.17 MiB	−5.25 MiB

Concurrency matrix

Concurrency	Base p50	HEAD p50	Delta	Base PSS	HEAD PSS	Delta	HEAD failures	Unstable B/H
1	19.86 ms	19.56 ms	−0.31 ms	751.00 KiB	794.00 KiB	+43.00 KiB	0	0/0
5	24.38 ms	23.45 ms	−0.94 ms	13.14 MiB	18.73 MiB	+5.58 MiB	0	0/0
10	22.27 ms	22.76 ms	+0.49 ms	23.18 MiB	27.89 MiB	+4.70 MiB	0	0/0
25	23.06 ms	22.59 ms	−0.47 ms	13.51 MiB	26.77 MiB	+13.26 MiB	0	0/0
100	22.44 ms	21.67 ms	−0.77 ms	20.67 MiB	13.33 MiB	−7.33 MiB	0	0/0

Frontend differential — ✅ 1,020/1,020 cases matched; 0 issues

Match	DOM mismatch	Diagnostic mismatch	Moli error	Reference error	Infrastructure error	Mismatched frames	Duration
1,020	0	0	0	0	0	0	7m 40.22s

Agent episodes · Moli vs Chromium — ✅ 8/8 Moli episodes passed; 0 failures

Fast contract	Episodes	Passed	Assertions	Ready p50	Episode p50
Moli	8	8	47/47	4.75 ms	25.73 ms
Chromium	8	8	47/47	278.22 ms	81.50 ms

Operation p50	Chromium	Moli	Delta	Delta %
navigate	44.18 ms	6.19 ms	−37.99 ms	−85.98%
observe	0.84 ms	0.86 ms	+0.02 ms	+2.82%
fill	3.68 ms	3.41 ms	−0.27 ms	−7.31%
click	5.05 ms	4.24 ms	−0.81 ms	−16.02%

Idle-resource episode	Peak RSS	Peak PSS	Average CPU	Peak CPU	Peak processes	Sampler
Moli	65.63 MiB	62.91 MiB	0.81%	12.00%	1	✅
Chromium	1304.98 MiB	334.86 MiB	2.13%	95.93%	13	⚠️ PSS partial

Runtime and CDP session contracts — ✅ 26 contract cases; 0 failures

Suite	Cases	Failures	Median case p50	Median case p95	Median peak PSS p50
Synthetic fetch	13	0	18.14 ms	19.02 ms	1.01 MiB
Long-lived CDP session	13	0	17.92 ms	18.23 ms	—

CDP smoke — ✅ 52/52 groups passed; 621 scenarios

Workers: 4 · cumulative group time: 153.98 s · failed groups: 0

Slowest group	Scenarios	Duration	Status
scrollbar-visibility	22	20.40 s	✅
target-lifecycle	24	12.46 s	✅
dom-input	27	10.19 s	✅
chromium-cdp	56	9.33 s	✅
multi-page	45	8.84 s	✅

WebMainBench · 545 pages — ✅ 545/545 completed; 0 unexpected failures

Completed	Successful	Expected DNS failure	Unexpected failures	Panics	Crashes	Timeouts	Empty output
545/545	544	1	0	0	0	0	0

Frozen HTML · Linux · --wait done · 45 seconds per page · no retries. Content quality is evaluated separately.

Diagnostics: webmainbench-results — every page's Markdown, stderr, and exit status. Source run and artifacts

All artifact fields are parsed by the trusted default-branch renderer; missing or invalid inputs remain visible as unavailable.

[github-actions]

Spider Bench A/B

✅ All benchmark browser service runs completed; results are informational.

Public HEAD: 184 / 240 rows; 38 / 48 sites produced rows; 10 unexpected empty sites.

✅ Deterministic fixture contract is clean.

Common ancestor 1acd30dde674 → HEAD 1f99b9fee9dd; benchmark order: head-first; workflow: success.

Public 48-site run · informational

Metric	Base	HEAD	Δ
Run status	success	success	—
Extracted rows / contract	184 / 240	184 / 240	0 (0.00%)
Contract fill	76.67%	76.67%	0.00 pp (0.00%)
Sites with rows	38 / 48	38 / 48	0 (0.00%)
Sites meeting row contract	36 / 48	36 / 48	0 (0.00%)
Unexpected empty sites	10	10	0 (0.00%)
Partial-row sites	2	2	0 (0.00%)
Site outcomes	extracted: 36; HTTP error + rows: 2; empty: 7; HTTP error + empty: 1; snapshot error + empty: 2	extracted: 37; HTTP error + rows: 1; empty: 7; HTTP error + empty: 1; snapshot error + empty: 2	—
Observed duration	46.97 s	55.05 s	+8.09 s (+17.22%)
Estimated CPU	49.30 s	48.65 s	-0.64 s (-1.31%)
Peak PSS	258.39 MiB	270.20 MiB	+11.81 MiB (+4.57%)
Peak RSS	262.07 MiB	272.86 MiB	+10.80 MiB (+4.12%)
Peak threads	25	25	0 (0.00%)
Mismatched sites	0	0	0
Timeout with items	0	0	0

Category rows	Base	HEAD	Δ
news	35 / 40	35 / 40	0
stocks	20 / 40	20 / 40	0
tech	35 / 40	35 / 40	0
sports	32 / 40	32 / 40	0
games	37 / 40	37 / 40	0
life	25 / 40	25 / 40	0

Public-site content, timing, and memory are noisy. This single A/B run reports evidence only; site outcome counts explain missing rows without treating them as a deterministic regression.

CPU and memory timelines

These are bounded 40-point views of the same complete process-tree samples used by the HTML report. CPU uses 100% per occupied logical core; memory lines are RSS first and PSS second when complete PSS samples are available.

Base

xychart-beta
    title "Base CPU"
    x-axis "Elapsed seconds" [1.038, 1.539, 2.54, 4.043, 5.044, 6.545, 8.046, 9.047, 10.549, 12.051, 13.052, 14.553, 15.554, 17.056, 18.558, 19.56, 21.063, 22.064, 23.566, 25.067, 26.068, 26.569, 27.569, 28.571, 30.073, 31.574, 32.574, 34.075, 35.578, 36.579, 38.081, 39.082, 40.582, 42.083, 43.085, 44.585, 45.586, 46.588]
    y-axis "CPU percent" 0 --> 200
    line [3.99, 93.94, 5.99, 129.72, 153.98, 185.97, 139.94, 173.56, 151.84, 61.95, 77.97, 147.95, 127.87, 113.88, 137.6, 23.94, 23.94, 97.93, 43.93, 113.85, 59.93, 0, 101.9, 141.87, 145.77, 133.79, 24.01, 153.65, 147.81, 101.87, 87.89, 51.94, 140.1, 19.97, 21.96, 24, 137.87, 189.73]

Loading

xychart-beta
    title "Base memory: RSS then PSS"
    x-axis "Elapsed seconds" [0.037, 1.038, 1.539, 2.54, 4.043, 5.044, 6.545, 8.046, 9.047, 10.549, 12.051, 13.052, 14.553, 15.554, 17.056, 18.558, 19.56, 21.063, 22.064, 23.566, 25.067, 26.068, 26.569, 27.569, 28.571, 30.073, 31.574, 32.574, 34.075, 35.578, 36.579, 38.081, 39.082, 40.582, 42.083, 43.085, 44.585, 45.586, 46.588, 46.968]
    y-axis "Memory MiB" 0 --> 500
    line [12.59, 62.14, 95.3, 101.05, 134.24, 152.77, 165.7, 172.97, 179.29, 184.8, 171.72, 161.11, 184.52, 190.52, 154.98, 182.59, 149.12, 150.12, 149.49, 175.43, 150.59, 150.63, 150.63, 262.07, 199.41, 215.19, 196.04, 167.46, 192.36, 170.34, 158.57, 153.11, 165.84, 233.77, 190.24, 175.82, 164.84, 190, 202.87, 0]
    line [9.39, 58.61, 91.81, 97.5, 130.72, 149.24, 162.27, 169.59, 178.21, 181.28, 168.17, 157.56, 180.99, 187.16, 151.43, 179.01, 145.33, 146.6, 145.95, 171.8, 146.62, 147.05, 147.05, 258.39, 195.87, 211.82, 192.64, 164.03, 188.36, 166.85, 155.17, 149.7, 162.48, 229.34, 186.84, 172.41, 161.45, 186.59, 199.4, 0]

Loading

HEAD

xychart-beta
    title "HEAD CPU"
    x-axis "Elapsed seconds" [1.539, 3.04, 4.542, 6.043, 7.045, 7.546, 9.046, 10.548, 12.05, 14.053, 15.553, 17.054, 18.556, 20.058, 21.559, 23.061, 24.563, 26.065, 27.567, 29.069, 30.57, 31.071, 32.072, 33.573, 35.075, 36.577, 38.079, 39.581, 41.583, 43.086, 44.589, 46.091, 47.593, 49.095, 50.595, 51.596, 52.097, 53.6]
    y-axis "CPU percent" 0 --> 500
    line [129.92, 134.07, 129.82, 79.81, 0, 2, 173.86, 95.92, 51.96, 101.75, 156.1, 133.91, 4, 2, 99.93, 107.92, 25.98, 117.67, 104.05, 155.8, 99.7, 107.88, 133.94, 117.78, 145.92, 161.91, 93.83, 107.84, 45.9, 163.62, 11.97, 0, 2, 37.94, 20.01, 205.56, 55.87, 147.83]

Loading

xychart-beta
    title "HEAD memory: RSS then PSS"
    x-axis "Elapsed seconds" [0.037, 1.539, 3.04, 4.542, 6.043, 7.045, 7.546, 9.046, 10.548, 12.05, 14.053, 15.553, 17.054, 18.556, 20.058, 21.559, 23.061, 24.563, 26.065, 27.567, 29.069, 30.57, 31.071, 32.072, 33.573, 35.075, 36.577, 38.079, 39.581, 41.583, 43.086, 44.589, 46.091, 47.593, 49.095, 50.595, 51.596, 52.097, 53.6, 55.054]
    y-axis "Memory MiB" 0 --> 500
    line [12.5, 115.21, 127.61, 136.63, 166.73, 147.64, 146.56, 188.55, 171.17, 176.61, 180.46, 165.28, 162.11, 147.75, 148.77, 144.18, 159.7, 144.71, 152.11, 194.48, 161.91, 217.27, 272.86, 201.65, 222.38, 152.27, 167.76, 201.25, 157.86, 154.2, 213.77, 170.97, 169.82, 169.81, 176.01, 167.14, 178.28, 180.12, 187.21, 0]
    line [9.22, 111.51, 123.49, 132.91, 162.98, 143.89, 142.81, 184.82, 168.33, 172.81, 176.79, 161.59, 158.27, 143.95, 144.98, 140.41, 155.9, 140.92, 148.31, 191.12, 158.17, 213.51, 270.2, 197.87, 218.63, 148.56, 164.74, 197.53, 154.15, 150.4, 210.59, 167.17, 166.16, 166.15, 172.35, 163.48, 175.24, 176.56, 183.71, 0]

Loading

Deterministic fixture · required contract

The fixture exercises 15 routes: 8 are expected to emit rows and 7 intentionally exercise empty, loading, or timeout behavior.

Metric	Base	HEAD	Δ
Run status	success	success	—
Extracted rows / contract	40 / 40	40 / 40	0 (0.00%)
Contract fill	100.00%	100.00%	0.00 pp (0.00%)
Sites with rows	8 / 15	8 / 15	0 (0.00%)
Sites meeting row contract	15 / 15	15 / 15	0 (0.00%)
Unexpected empty sites	0	0	0
Partial-row sites	0	0	0
Site outcomes	extracted: 7; HTTP error + rows: 1; empty: 3; timeout + empty: 4	extracted: 7; HTTP error + rows: 1; empty: 3; timeout + empty: 4	—
Observed duration	19.87 s	19.89 s	+0.02 s (+0.12%)
Estimated CPU	0.43 s	0.44 s	+0.01 s (+1.85%)
Peak PSS	66.32 MiB	66.42 MiB	+0.11 MiB (+0.16%)
Peak RSS	69.91 MiB	70.07 MiB	+0.16 MiB (+0.23%)
Peak threads	18	18	0 (0.00%)
Mismatched sites	0	0	0
Timeout with items	0	0	0

Full HTML, JSON, CSV, logs, and page snapshots: workflow run and full spider-bench-results artifact.

The public-web A/B is informational. The exact HEAD fixture contract runs as a separate required CI check; timing and public-site content deltas remain non-blocking.

[github-actions]

Sequential Navigation Soak A/B

✅ HEAD completed the 200-navigation resilience and memory observation.

One browser process, one target, and one CDP session navigate CSDN → SegmentFault → Huaban → example.com repeatedly. This run issued 200 Page.navigate commands per binary.

Common ancestor 1acd30dde6 → HEAD 1f99b9fee9; benchmark order: head-first; workflow: success.

Session resilience

Metric	Base	HEAD
Attempted / planned	200 / 200	200 / 200
Direct observable pass	197 / 200	200 / 200
Failures / recovered	3 / 3	0 / 0
Recovery failures	0	0
Lifecycle order violations	0	0
Network order violations	0	0
Wall time	642.8 s	544.7 s

A public-page failure is reported separately from an unrecoverable session. The soak requires all navigation attempts, zero failed recovery, zero lifecycle/network ordering violations, and complete resource evidence.

Process-tree memory

Metric	Base	HEAD	Δ
Boundary samples	200	200	0
Boundary RSS samples	200	200	0
Boundary PSS samples	200	200	0
Periodic peak RSS	264.2 MiB	266.0 MiB	+1.8 MiB
Periodic peak PSS	261.4 MiB	262.5 MiB	+1.2 MiB
RSS · first 10 avg	184.6 MiB	176.6 MiB	-8.0 MiB
RSS · last 10 avg	187.4 MiB	181.1 MiB	-6.3 MiB
RSS · last − first window	+2.8 MiB	+4.6 MiB	+1.8 MiB
RSS · warm slope / 100 nav	+8.8 MiB	+3.1 MiB	-5.7 MiB
PSS · first 10 avg	181.3 MiB	173.2 MiB	-8.1 MiB
PSS · last 10 avg	184.4 MiB	177.7 MiB	-6.7 MiB
PSS · last − first window	+3.1 MiB	+4.5 MiB	+1.4 MiB
PSS · warm slope / 100 nav	+8.9 MiB	+3.3 MiB	-5.6 MiB
Peak file descriptors	61	63	2
Peak threads	31	31	0

HEAD memory by 50-navigation quarter

HEAD range	Avg RSS	Final RSS	Peak RSS	Avg PSS	Final PSS	Peak FD
1–50	179.0 MiB	157.7 MiB	239.4 MiB	175.4 MiB	154.0 MiB	52
51–100	186.4 MiB	178.4 MiB	233.2 MiB	182.9 MiB	174.9 MiB	54
101–150	190.7 MiB	166.7 MiB	258.0 MiB	187.3 MiB	163.6 MiB	54
151–200	187.8 MiB	140.7 MiB	242.3 MiB	184.3 MiB	137.3 MiB	53

Raw reports and per-navigation boundary samples: workflow run and full sequential-navigation-soak-results artifact.

Public-site timing and memory are observational. Use the A/B deltas and trend shape as evidence, not as a deterministic performance threshold.

[lanyue-llk]

Closing as obsolete. This writes an earlier benchmark snapshot into the product README; the current report and evidence are owned by browser-eval and the historical numbers are no longer the publication source.

[lanyue-llk]

Reopened after correcting the consolidation scope: this is an independent pre-existing stack and should not have been closed as part of #1501 cleanup. Its original scope and base relationship are preserved; this does not add it to #1501.