Every raw model response is publicly verifiable
Download image (with QR code)Share on XShare on LinkedIn

WeChat: save the image and long-press the QR code in a chat.

Diagnose my brand — free

From memory. For the record.

The certificate is for sale.
The rank is not.

The same question in two languages — once in Chinese, once in English — to the same panel of mainstream LLMs, on the same day, with both rankings side by side. Change the language and the list changes. Every raw response stays on file for anyone to check, including the ones that never answered.

See today's rankingsDiagnose my brand — free

For brands re-measured on an agreed schedule: AI Visibility Monitoring →

Verified record—

Pet Dryer Box / Dog Washer

Flying Pig Grooming65.18/ 100

Ranked 1st of 54 candidate brands

2026-10-1812 models

Sample layout, built from the most recent evaluation — real brand, real score, real per-model ranks. A certificate number is issued at purchase; this brand has not bought one.

Behind every ranking
27
Rankings published
1,207
Raw model responses kept on file
12
Models queried per ranking

These aren't marketing figures — every one of those responses is stored verbatim and reprinted on the certificate page, so you can open any ranking and check the arithmetic yourself. How we compute it →

How a ranking is produced

Three steps. All three are logged, and you can check any of them.

  1. 01

    The same question, in two languages

    One question goes to every model on the panel in the same run, then the same question goes out again in the other language, on the same day. No follow-up prompts, no re-rolls, and no discarding an answer we don't like.

  2. 02

    Normalise across languages, then weight

    The answers come back in different languages and spellings. Brands are merged into one canonical name, then each model's ranking is weighted by its panel weight and aggregated into a single index.

  3. 03

    Keep every raw answer

    Each response is stored verbatim alongside its model id, weight and contribution — and reprinted in full on the verification page of every certificate.

Evaluation panel · 12 mainstream LLMs currently in use
openai/gpt-5.6-lunastealth/space-bunny-alphaz-ai/glm-5.3-flashxiaomi/mimo-v2.6-flash
Show 8 more models
qwen/qwen3.8-max-0902google/gemini-3.7-flashdeepseek/deepseek-v4-flashnvidia/nemotron-3-ultra-550b-a55b:freex-ai/grok-4.7stepfun/step-5-previewanthropic/claude-sonnet-5moonshotai/kimi-k3
Read the full methodology, including when the number shouldn't be trusted

Hot Topics

The same method, pointed at a question people are asking right now.

Observed 2026-09-02

AI Glasses

1Rokid66.52
2XREAL64.40
3Meta59.39
4Ray-Ban52.33

2026-09-0212 models4 of 30 candidates shownon the board 2026-09-02

View full ranking →
Observed 2026-09-03

LLM API Plans

1OpenAI100.00
2Anthropic90.00
3Google46.96
4Mistral AI44.84

2026-09-0312 models4 of 29 candidates shownon the board 2026-09-03

View full ranking →
Observed 2026-09-22

AI Coding Agents

1GitHub Copilot100.00
2Cursor90.00
3Windsurf (Devin Desktop)74.86
4Claude49.29

2026-09-2212 models4 of 19 candidates shownon the board 2026-09-22

View full ranking →

Two measurement lines

One panel, two readings. One ranks the head of a category; the other records whether a name shows up at all. The two numbers are not interchangeable and are never shown side by side.

Brand index0–100

The twelve seats' lists are weighted into a single index, then ranked.

The six weekly questions are re-run automatically every week, which suits watching the head of a category move; other rankings are re-run on demand.

Weekly measurements →

Mention rate%

The share of answers in one batch that name a given product, each row with a Wilson 95% interval.

One run splits a category into several scenarios and phrasings, so it suits long-tail names: where they show up and where they do not.

Scenario collections →

Weekly monitorBrand index

Re-measured automatically every Monday: three questions, each asked once in Chinese and once in English, six cards in matched pairs — each card is one dated measurement.

Smart Pet Feeder / Water Dispenser (asked in Chinese)

1PETKIT106.20
2Xiaomi57.79
3PetSafe52.98
4PETLIBRO51.92

2026-10-1812 models4 of 37 candidates shown

View full ranking →
pet-hardware

Smart Pet Feeder / Water Dispenser

1PETLIBRO87.78
2PetSafe79.71
3WOPET71.07
4PETKIT66.22

2026-10-1812 models4 of 31 candidates shown

View full ranking →
pet-hardware

Smart Cat Litter Box (asked in Chinese)

1PETKIT91.50
2Litter-Robot80.26
3CATLINK69.69
4Neakasa35.91

2026-10-1812 models4 of 43 candidates shown

View full ranking →
pet-hardware

Smart Cat Litter Box

1PETKIT90.92
2Litter-Robot78.79
3PetSafe61.69
4CatGenie39.12

2026-10-1812 models4 of 30 candidates shown

View full ranking →
pet-hardware

Pet Dryer Box / Dog Washer (asked in Chinese)

1PETKIT74.49
2Homerunpet66.39
3Shernbao31.30
4Double K14.83

2026-10-1812 models4 of 80 candidates shown

View full ranking →
pet-hardware

Pet Dryer Box / Dog Washer

1Flying Pig Grooming65.18
2Shernbao52.63
3Aeolus24.84
4iPettie24.75

2026-10-1812 models4 of 52 candidates shown

View full ranking →

AI toolingBrand index

For readers who evaluate AI engineering. This group is not on the weekly schedule: it is re-run on demand with exactly the same method as every other ranking, and the date on each card is the date of that measurement.

ai-tooling

Most Popular Embedding Model API Brands

1OpenAI100.00
2Cohere83.65
3Hugging Face54.47
4Voyage AI50.35

2026-08-3112 models4 of 27 candidates shown

View full ranking →
ai-tooling

Most Popular LLM Observability Platforms

1LangSmith87.93
2Weights & Biases68.32
3Arize AI64.65
4Helicone59.29

2026-08-3112 models4 of 29 candidates shown

View full ranking →
ai-tooling

Most Popular AI Model Gateway / Routing Platforms

1OpenRouter91.52
2LiteLLM85.17
3Portkey66.73
4Helicone43.59

2026-08-3112 models4 of 38 candidates shown

View full ranking →
ai-tooling

Most Popular Code Execution Sandboxes for AI Agents

1E2B86.98
2Docker75.25
3Modal69.41
4Replit51.00

2026-08-3112 models4 of 41 candidates shown

View full ranking →
ai-tooling

Most Popular Browser Automation Frameworks for AI Agents

1Playwright99.25
2Selenium85.75
3Puppeteer85.00
4Cypress45.36

2026-08-3112 models4 of 34 candidates shown

View full ranking →
ai-tooling

Most Popular Web Search APIs for AI Agents

1Google84.75
2Tavily80.79
3Bing77.03
4SerpApi76.26

2026-08-3112 models4 of 25 candidates shown

View full ranking →
ai-tooling

Most Popular LLM Application / Agent Development Frameworks

1LangChain100.00
2LlamaIndex83.76
3AutoGen61.64
4CrewAI56.02

2026-08-3112 models4 of 29 candidates shown

View full ranking →
ai-tooling

Most Popular Vector Database Brands

1Pinecone96.90
2Milvus92.75
3Weaviate76.57
4Chroma67.67

2026-08-3112 models4 of 17 candidates shown

View full ranking →

Scenario collectionsMention rate

One model panel, one day, one category split into several scenario questions. Each scenario is its own cell with its own candidate pool, so the cells are not merged into a single ranking.

AI coding agents · scenario rankings

Measured 2026-09-22 · 12-seat model panel · 10 scenarios · methodology #2 v2026-09-22.1

View all →
5 more archived rankings (no longer updated; links and verification pages stay live) →
Most Popular Electric Toothbrush Brands WorldwideMost Popular Smart Lock BrandsMost Popular Desktop Diffuser/Humidifier BrandsMost Popular Desktop Mosquito Lamp BrandsMost Popular Electric Vehicle Brands in Hong Kong