ML platforms
Asked as “machine learning platform”, and as “MLOps platform for training and deploying models”, on behalf of a mid-market B2B company. 58 first choices recorded across the direct, paraphrase, budget and scale prompts, fourteen models each.
Standing · first-choice share
24%
Contested · Azure Machine Learning 19%
24Databricks Mosaic AI19Azure Machine Learning09Dataiku48others
24% of first choices, contested.
Since October 2026↗new leaderNew leader since October 2026: Databricks Mosaic AI (24%) replaces Azure Machine Learning (21% then, 19% now), 5 points clear, inside the 11-point floor.Databricks Mosaic AI leads at 24%, replacing Azure Machine Learning, which led at 21% and stands at 19% now: 5 points clear, inside the floor, so the swap reads as unsettled.
By buyer segment
The same question asked on behalf of a different buyer. Each standing is computed within its segment; they sit side by side and are never added together.
Small business
Google Vertex AI20%Contested▲+1Since October 2026: 19% → 20%, +1 point. Inside the 11-point floor: within noise.Standing →Mid-market
Databricks Mosaic AI24%Contested↗new leaderNew leader since October 2026: Databricks Mosaic AI (24%) replaces Azure Machine Learning (21% then, 19% now), 5 points clear, inside the 11-point floor.This pageEnterpriseDatabricks Mosaic AI49%Clear leader▲+5Since October 2026: 44% → 49%, +5 points. Inside the 11-point floor: within noise.Standing →
Google Vertex AI20%Contested▲+1Since October 2026: 19% → 20%, +1 point. Inside the 11-point floor: within noise.Standing →Mid-marketDatabricks Mosaic AI24%Contested↗new leaderNew leader since October 2026: Databricks Mosaic AI (24%) replaces Azure Machine Learning (21% then, 19% now), 5 points clear, inside the 11-point floor.This pageEnterpriseDatabricks Mosaic AI49%Clear leader▲+5Since October 2026: 44% → 49%, +5 points. Inside the 11-point floor: within noise.Standing →The standing
Share is the count of first choices across the direct, paraphrase, budget and scale prompts, over all fourteen models, for a mid-market B2B company. Ordered by share.
ProductFirst-choice shareNegative rateLabelsQuadrantSince October 2026
01Databricks Mosaic AI24%9%53accepted challenger▲+10Since October 2026: 14% → 24%, +10 points. Inside the 11-point floor: within noise.14% → 24%
Databricks Mosaic AI24%9%53accepted challenger▲+10Since October 2026: 14% → 24%, +10 points. Inside the 11-point floor: within noise.14% → 24%02
Azure Machine Learning19%12%58accepted challenger▼−2Since October 2026: 21% → 19%, −2 points. Inside the 11-point floor: within noise.21% → 19%
Azure Machine Learning19%12%58accepted challenger▼−2Since October 2026: 21% → 19%, −2 points. Inside the 11-point floor: within noise.21% → 19%03
Dataiku9%0%23accepted challenger▲+9Since October 2026: 0% → 9%, +9 points. Inside the 11-point floor: within noise.0% → 9%
Dataiku9%0%23accepted challenger▲+9Since October 2026: 0% → 9%, +9 points. Inside the 11-point floor: within noise.0% → 9%04
Amazon SageMaker7%18%67accepted challenger▼−4Since October 2026: 11% → 7%, −4 points. Inside the 11-point floor: within noise.11% → 7%
Amazon SageMaker7%18%67accepted challenger▼−4Since October 2026: 11% → 7%, −4 points. Inside the 11-point floor: within noise.11% → 7%05
MLflow7%9%23accepted challenger=heldSince October 2026: 7% → 7%, ±0 points. Inside the 11-point floor: within noise.7% → 7%
MLflow7%9%23accepted challenger=heldSince October 2026: 7% → 7%, ±0 points. Inside the 11-point floor: within noise.7% → 7%06
Snowflake Cortex AI3%7%15accepted challenger▲+3Since October 2026: 0% → 3%, +3 points. Inside the 11-point floor: within noise.0% → 3%
Snowflake Cortex AI3%7%15accepted challenger▲+3Since October 2026: 0% → 3%, +3 points. Inside the 11-point floor: within noise.0% → 3%07
PyTorch3%10%10accepted challenger▼−1Since October 2026: 4% → 3%, −1 point. Inside the 11-point floor: within noise.4% → 3%
PyTorch3%10%10accepted challenger▼−1Since October 2026: 4% → 3%, −1 point. Inside the 11-point floor: within noise.4% → 3%08
TensorFlow3%10%10accepted challenger▲+1Since October 2026: 2% → 3%, +1 point. Inside the 11-point floor: within noise.2% → 3%
TensorFlow3%10%10accepted challenger▲+1Since October 2026: 2% → 3%, +1 point. Inside the 11-point floor: within noise.2% → 3%09Google Vertex AI Lab in the set2%13%52accepted challenger▼−7Since October 2026: 9% → 2%, −7 points. Inside the 11-point floor: within noise.9% → 2%
Google Vertex AI Lab in the set2%13%52accepted challenger▼−7Since October 2026: 9% → 2%, −7 points. Inside the 11-point floor: within noise.9% → 2%10
DataRobot2%14%29accepted challenger▼−2Since October 2026: 4% → 2%, −2 points. Inside the 11-point floor: within noise.4% → 2%
DataRobot2%14%29accepted challenger▼−2Since October 2026: 4% → 2%, −2 points. Inside the 11-point floor: within noise.4% → 2%11
H2O.ai2%0%15accepted challenger▲+2Since October 2026: 0% → 2%, +2 points. Inside the 11-point floor: within noise.0% → 2%
H2O.ai2%0%15accepted challenger▲+2Since October 2026: 0% → 2%, +2 points. Inside the 11-point floor: within noise.0% → 2%Show the four products at 0%, ordered by negative rate
15
Kubeflow0%44%16criticized challenger=heldSince October 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise.0% → 0%
Kubeflow0%44%16criticized challenger=heldSince October 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise.0% → 0%14
Hugging Face0%20%10accepted challenger=heldSince October 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise.0% → 0%
Hugging Face0%20%10accepted challenger=heldSince October 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise.0% → 0%12
Domino Data Lab0%0%11accepted challenger=heldSince October 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise.0% → 0%
Domino Data Lab0%0%11accepted challenger=heldSince October 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise.0% → 0%13Weights & Biases0%0%10accepted challenger▼−2Since October 2026: 2% → 0%, −2 points. Inside the 11-point floor: within noise.2% → 0%
The floor is 11 points of share, measured: how far the models move a leader on their own when the same questions are asked twice with nothing changed. A larger change is movement; a smaller one can't be told apart from repeat variation, and both are shown. Each index measures its own floor from its own repeat, over its own categories, so the floor differs from one index to the next. Both editions asked the same fourteen models. How the floor is measured
Bars are the share of first choices, 0 to 100Every product with at least 10 labels here. Every product name links to its product page.
Google Vertex AI is made by Google, whose model Gemini 3.5 Flash is in the set. On the four share prompts that model made it the first choice zero times of 4; the other thirteen models one time of 52. Lab treatment is defined on the method page; the row is marked, not excluded.
Head to head
All twenty-eight head-to-head pages: the top eight products, each against each
Databricks Mosaic AI vs MLflowDatabricks Mosaic AI vs Snowflake Cortex AIDatabricks Mosaic AI vs PyTorchDatabricks Mosaic AI vs TensorFlowAzure Machine Learning vs DataikuAzure Machine Learning vs Amazon SageMakerAzure Machine Learning vs MLflowAzure Machine Learning vs Snowflake Cortex AIAzure Machine Learning vs PyTorchAzure Machine Learning vs TensorFlowDataiku vs Amazon SageMakerDataiku vs MLflowDataiku vs Snowflake Cortex AIDataiku vs PyTorchDataiku vs TensorFlowAmazon SageMaker vs MLflowAmazon SageMaker vs Snowflake Cortex AIAmazon SageMaker vs PyTorchAmazon SageMaker vs TensorFlowMLflow vs Snowflake Cortex AIMLflow vs PyTorchMLflow vs TensorFlowSnowflake Cortex AI vs PyTorchSnowflake Cortex AI vs TensorFlowPyTorch vs TensorFlow
Recommended versus criticized
Every product with at least 10 labels here, on both axes. The 30% line names a quadrant, not the verdict above: that one needs more than 40%.
Criticized challengerCriticized default
Negative label rate →
010203040506070809101112W13
1415Accepted challengerEndorsed leader
0%First-choice share → · lines at 30% share and 25% negative40%
Key
01Databricks Mosaic AI24%
Databricks Mosaic AI24%02Azure Machine Learning19%
Azure Machine Learning19%03Dataiku9%
Dataiku9%04Amazon SageMaker7%
Amazon SageMaker7%05MLflow7%
MLflow7%06Snowflake Cortex AI3%
Snowflake Cortex AI3%07PyTorch3%
PyTorch3%08TensorFlow3%
TensorFlow3%09Google Vertex AI2%
Google Vertex AI2%10DataRobot2%
DataRobot2%11H2O.ai2%
H2O.ai2%12Domino Data Lab0%
Domino Data Lab0%13Weights & Biases0%
14Hugging Face0%
Hugging Face0%15Kubeflow0%
Kubeflow0%What they warned about
Two of fourteen models held their first choice under the paraphrase. Claude Haiku 4.5, Gemini 3.5 Flash, Perplexity Sonar, Grok 4.20, Mistral Small, Llama 4 Maverick, Qwen 3.7 Flash, Kimi K2, GLM 4.7 FlashX, MiniMax M2.5, GPT-6 Luna and Muse Glimmer 30B changed. A high negative share on a product with few labels is a warning. A low share on a product with many labels is salience, not sentiment.
Amazon SageMaker18%
12 of 67 labels negative · 10 of 14 models · 1 hard negative
“Avoid heavy enterprise tools like Databricks or full SageMaker unless you have specific needs” Grok 4.20, budget prompt
Azure Machine Learning12%
7 of 58 labels negative · 7 of 14 models
“Azure ML can be compromised through device code phishing, where attackers steal access tokens and exfiltrate models stored in the platform” Muse Glimmer 30B, negative prompt
Kubeflow44%
7 of 16 labels negative · 6 of 14 models · 3 hard negative
“Avoid starting with Kubeflow unless the company already has a strong Kubernetes platform team” Perplexity Sonar, paraphrase prompt
Google Vertex AI13%
7 of 52 labels negative · 6 of 14 models
“Vertex AI and SageMaker have been flagged as potentially costly or overly complex for smaller organizations” Perplexity Sonar, negative prompt
What they cite
Fourteen of the fourteen models returned a source list in this edition: five return the pages they cited and nine the search results they consulted.
Sites the answers cite
64 of 84 answers in this category came back with a source list, from 14 of 14 models: citations where the model returns them, or the search results it consulted. 676 links across 272 sites, every framing counted. Ranked by the number of answers carrying the site or page. 54 of the 252 answers across every segment cited this index's own page for the category; the method page measures whether that reading tilts an answer.
vendor site · G231 answers · 44 citations · 11 models
22 answers · 22 citations · 10 models
20 answers · 23 citations · 8 models
publisher (Gane)16 answers · 18 citations · 8 models
vendor site · LinkedIn12 answers · 13 citations · 5 models
10 answers · 15 citations · 9 models
9 answers · 10 citations · 7 models
9 answers · 10 citations · 6 models
9 answers · 9 citations · 8 models
vendor site · Google9 answers · 9 citations · 5 models
vendor site · Microsoft8 answers · 9 citations · 5 models
8 answers · 8 citations · 7 models
Pages the answers cite
The ten pages named in the most answers, by full address. A page here is one the models returned with a recommendation, not one the index endorses.
12 answers · 8 models
12 answers · 8 models
11 answers · 7 models
10 answers · 6 models
9 answers · 8 models
8 answers · 7 models
8 answers · 7 models
8 answers · 5 models
8 answers · 5 models
7 answers · 7 models
Search against answers
Each company's standing in the answers beside its site's footprint in Google search, one row a site: the products the models named on it with their shares, and the share they add up to; monthly searches on Google, and DataForSEO's estimate of AI search demand (modeled from search signals, directional, not a count of queries to any assistant), for the most-searched of the company's and its products' names (the name is in each row's hover text); estimated monthly organic visits to the site; and its best position in Google's top ten for “best machine learning platform”, “machine learning platform”, “machine learning platforms”. US estimates from DataForSEO and Google's Ads Transparency Center. A small company's site, or a mid-sized company's site for its flagship, is marked company; a product on a large parent's site (Google, Microsoft) has no site figures. A column with no figures for this category is left out, and an empty cell means none were seen, not none exist. Two measurements side by side: neither is read as the cause of the other.
| Company and products | Share | Own site cited | Name searches, Google | AI search demand, est. | Organic visits | Paid search a month, est. | Google ads, last 30 days |
|---|---|---|---|---|---|---|---|
| Databricks databricks.com Databricks Mosaic AI 24% | 24% | 125 | 135,000 | 8,368 | 689,756 company | $34,664 | 120 |
| Azure parent site Azure Machine Learning 19% | 19% | 148 | 301,000 | 47,904 | |||
| Dataiku dataiku.com | 9% | 74 | 9,900 | 266 | 12,714 | 93 | |
| Amazon parent site Amazon SageMaker 7% | 7% | 159 | 151,000,000 | 1,380,939 | |||
| MLflow mlflow.org | 7% | 58 | 9,900 | 438 | 6,498 | 1 | |
| Google Colab colab.research.google.com | 7% | 18 | 301,000 | 900 | 240,528 | ||
| KNIME Analytics Platform knime.com | 5% | 24 | 320 | 3 | 28,171 | 23 | |
| Snowflake Cortex AI snowflake.com | 3% | 28 | 2,900 | 87 | 718,749 | 118 | |
| PyTorch pytorch.org | 3% | 11 | 40,500 | 4,030 | 150,980 | ||
| TensorFlow tensorflow.org | 3% | 16 | 18,100 | 2,745 | 198,535 | 3 |
Google's top ten for the three searches hold 22 sites; 1 of them are among the sites the models cited here (comidor.com). The first result for “best machine learning platform” is anaconda.com.
How they answered
Six framings of the same buying question, each sent to every model in a fresh session with search on. One row per model, so a row shows whether it held its answer under rewording, what it named when cost was the constraint, and what it argued against. Computed from the raw judge labels.ShowHide
How they answered
Six framings of the same buying question, each sent to every model in a fresh session with search on. One row per model, so a row shows whether it held its answer under rewording, what it named when cost was the constraint, and what it argued against. Computed from the raw judge labels.
ShowHideModelDirect“What is the best machine learning platform for a mid-market B2B company?”Paraphrase“Which MLOps platform for training and deploying models would you recommend to a mid-sized B2B company?”Comparative“What are the top machine learning platforms and how do they differ?”Budget-constrained“What is the best machine learning platform for a company with a limited budget?”Scale-constrained“We are a 500 person company evaluating a machine learning platform. What should we look at?”Negative“Which machine learning platforms should I avoid or be cautious about?”
Claude Haiku 4.5no first choiceDatabricks Mosaic AI, MLflowChangedno first choicePyTorch, TensorFlowno first choicenothing named
Seven alternatives
Amazon SageMaker, Azure Machine Learning, Clearly.ai, DataRobot, Databricks Mosaic AI, Google Vertex AI, H2O AutoMLTwo alternatives
Amazon SageMaker, Google Vertex AITwo alternatives
Google Colab, scikit-learnGPT-5.4 miniDatabricks Mosaic AIDatabricks Mosaic AIHeldagainst: Kubeflow, MLflowno first choiceGoogle Colab, Kaggleagainst: Amazon SageMakerno first choiceagainst: OpenAI
Two alternatives
Amazon SageMaker, Snowflake Cortex AITwo alternatives
Amazon SageMaker, Google Vertex AISix alternatives
Amazon SageMaker, Azure Machine Learning, Databricks Mosaic AI, Google Vertex AI, MLflow, Microsoft FabricFour alternatives
Azure Machine Learning, Google Vertex AI, MLflow, PyTorchGemini 3.5 FlashDataikuDatabricks Mosaic AIChangedagainst: KubeflowAmazon SageMakerMLflowagainst: Anthropic, Google Cloud AutoML & Vertex AI, OpenAIno first choiceagainst: Amazon SageMaker, Azure Machine Learning, Databricks Mosaic AI, GCP Vertex AI, Snowflake Cortex AI
Six alternatives
Amazon SageMaker, DataRobot, Databricks Mosaic AI, GCP Vertex AI, Google BigQuery ML, Snowflake Cortex AISeven alternatives
Amazon SageMaker, BentoML, ClearML, GCP Vertex AI, Modal, Outerbounds, Weights & BiasesFour alternatives
Azure Machine Learning, DataRobot, Databricks Lakehouse Platform, Google Vertex AISixteen alternatives
AWS SageMaker Studio Lab, Amazon SageMaker, DVC, DeepInfra, Google Colab, Groq, H2O-3, Hugging Face, Hugging Face Spaces, Kaggle, Ollama, PyCaret, PyTorch, Together AI, scikit-learn, vLLMPerplexity SonarAzure Machine LearningAmazon SageMaker, Azure Machine Learning, Google Vertex AIChangedagainst: Kubeflowno first choiceXGBoost, scikit-learnno first choiceagainst: Amazon SageMaker, Azure Machine Learning, Google Vertex AI, Kubeflow, Ray-based systems
Three alternatives
Amazon SageMaker, Databricks Mosaic AI, DataikuThree alternatives
Databricks Mosaic AI, Domino Data Lab, MLflowNine alternatives
Amazon SageMaker, Azure Machine Learning, DataRobot, Databricks Mosaic AI, Dataiku, Domino Data Lab, Google Vertex AI, H2O.ai, Hugging FaceSeven alternatives
Amazon SageMaker, Deepnote, Google Vertex AI, KNIME Analytics Platform, Paperspace, PyTorch, TensorFlowGrok 4.20Dataikuagainst: DataRobot, MLflowDatabricks Mosaic AIChangedAmazon SageMakerGoogle Colabagainst: Amazon SageMaker, DataRobot, Databricks Mosaic AIno first choiceagainst: Amazon SageMaker, Anthropic's Claude, Azure Machine Learning, ChatGPT, Google Gemini, Google Vertex AI, Microsoft Copilot, Slack GPT
Four alternatives
Amazon SageMaker, Azure Machine Learning, Databricks Mosaic AI, Google Vertex AISeven alternatives
Amazon SageMaker, Azure Machine Learning, DataRobot, Dataiku, Google Vertex AI, Kubeflow, MLflowFour alternatives
Azure Machine Learning, Databricks Mosaic AI, Dataiku, Google Vertex AIFive alternatives
Apache MLflow, Azure Machine Learning, Hugging Face, Kubeflow, TensorFlowMistral SmallAzure Machine Learning, Snowflake Cortex AIDatabricks Mosaic AIChangedno first choiceOpenAI Platformno first choiceagainst: Amazon SageMaker, Azure Machine Learning, ChatGPT, DagsHub, DeepSeek, Google Vertex AI, Kubeflow
Three alternatives
DataRobot, Databricks Mosaic AI, Google Vertex AIFour alternatives
Amazon SageMaker, Azure Machine Learning, DataRobot, Google Vertex AIFour alternatives
Amazon SageMaker, Google Vertex AI, KNIME Analytics Platform, MLflowDeepSeek V4 FlashAzure Machine LearningAzure Machine LearningHeldagainst: Google Vertex AIno first choiceGoogle Colab, H2O.aiagainst: Amazon SageMakerno first choiceagainst: Amazon SageMaker, Azure Machine Learning, DataRobot, Google Vertex AI, Grok, H2O Driverless AI, Hugging Face, ModelScope, OpenCSG, RapidMiner
Four alternatives
Amazon SageMaker, DataRobot, Databricks Mosaic AI, Snowflake Cortex AIThree alternatives
Amazon SageMaker, Databricks Mosaic AI, MLflowThree alternatives
Azure Machine Learning, Google Vertex AI, KNIME Analytics PlatformLlama 4 MaverickAzure Machine Learningagainst: Databricks Mosaic AIDatabricks Mosaic AIChangedno first choiceno first choiceno first choiceagainst: Amazon SageMaker, DagsHub, Kubeflow
Four alternatives
DataRobot, Google Vertex AI, IBM watsonx.ai, Snowflake Cortex AIFour alternatives
Amazon SageMaker, Dataiku, Domino Data Lab, Google Vertex AIQwen 3.7 FlashDataRobotDatabricks Mosaic AIChangedAmazon SageMakerDataiku, KNIME Analytics Platformno first choiceagainst: Databricks Mosaic AI, OpenAI, SAS
Three alternatives
Amazon SageMaker, Azure Machine Learning, Databricks Mosaic AIThree alternatives
Amazon SageMaker, Azure Machine Learning, Weights & BiasesFive alternatives
Azure Machine Learning, DataRobot, Databricks Mosaic AI, Google Vertex AI, RapidMinerFour alternatives
Amazon SageMaker, Azure Machine Learning, Google Vertex AI, Hugging FaceKimi K2Azure Machine Learning, Dataikuagainst: Databricks Mosaic AI, Palantir FoundryDatabricks Mosaic AIChangedagainst: Kubeflowno first choicePyTorch, TensorFlowno first choiceagainst: AWS Lambda, Amazon SageMaker, Azure Machine Learning, Google Cloud AutoML, Google Cloud Functions, Google Vertex AI
One alternative
Amazon SageMakerFive alternatives
Amazon SageMaker, Azure Machine Learning, Domino Data Lab, Google Vertex AI, Weights & BiasesThree alternatives
Amazon SageMaker, Azure Machine Learning, Google Vertex AIGLM 4.7 FlashXAmazon SageMaker, Azure Machine Learningagainst: DataRobotDatabricks Mosaic AIChangedno first choiceGoogle Colabno first choiceagainst: Amazon SageMaker, Anthropic Claude Enterprise, ChatGPT, Civitai, GitHub-hosted model repositories, Google Gemini, Hugging Face, TensorFlow Playground
Two alternatives
Databricks Mosaic AI, Google Vertex AIFour alternatives
Amazon SageMaker, Azure Machine Learning, DataRobot, DataikuFive alternatives
Alteryx Machine Learning, Amazon SageMaker, Azure Machine Learning, Dataiku, H2O.aiMiniMax M2.5Databricks Mosaic AI, Snowflake Cortex AIAmazon SageMaker, Azure Machine LearningChangedno first choiceKNIME Analytics PlatformAmazon SageMaker, Databricks Mosaic AInothing named
Five alternatives
Amazon SageMaker, Azure Machine Learning, Dataiku, Domino Data Lab, Google Vertex AIFive alternatives
DataRobot, Databricks MLflow, Domino Data Lab, Google Vertex AI, Kubeflow on-premises / AnthosThree alternatives
Google Colab, Google Vertex AI, MLflowSix alternatives
Azure Machine Learning, Dataiku, Google Vertex AI, H2O.ai, MLflow, Weights & BiasesGPT-6 LunaDatabricks Mosaic AIagainst: SageMaker AIAzure Machine LearningChangedAmazon SageMaker, Azure Machine Learning, Gemini Enterprise Agent PlatformMLflowno first choicenothing named
Three alternatives
Azure Machine Learning, Gemini Enterprise Agent Platform, Snowflake Cortex AITwo alternatives
Amazon SageMaker, Databricks Mosaic AI with MLflowFour alternatives
Databricks Mosaic AI, Kubeflow, MLflow, Snowflake Cortex AITwo alternatives
Amazon SageMaker, Hugging FaceMuse Glimmer 30BAzure Machine Learning, DataikuDatabricks Mosaic AI, MLflowChangedno first choiceKNIME Analytics Platformno first choiceagainst: PyTorch, TensorFlow, scikit-learnagainst: Amazon SageMaker, Azure Machine Learning, BigML, ChatGPT, Claude, DagsHub, Deep Lake, Google Gemini, Google Vertex AI, Mage, Vanna.AI, Weave, ZenML
Four alternatives
Amazon SageMaker, Databricks Mosaic AI, Google Vertex AI, Snowflake Cortex AIThree alternatives
Amazon SageMaker, Azure Machine Learning, Google Vertex AIThree alternatives
Anaconda, Google Colaboratory, TensorFlowSix alternatives
Amazon SageMaker, Azure Machine Learning, Databricks Mosaic AI, Dataiku, Google Vertex AI, H2O.aiBold is the first choiceAlternatives are counted; the count opens them.What the answer argued against
The record
One row per call: the version string exactly as returned, whether the model searched, sources cited, and latency.
Eighty-four rows: every prompt, every model, every answer.
| Prompt | Model | Version string | Time (UTC) | Searched | Sources | Latency |
|---|---|---|---|---|---|---|
| Direct recommendation | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 11:15 | no | 0 | 5 s |
| Direct recommendation | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 11:54 | yes | 3 | 5 s |
| Direct recommendation | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 09:50 | yes | 11 | 23 s |
| Direct recommendation | Perplexity Sonar | sonar | 2026-11-01 11:28 | yes | 10 | 3 s |
| Direct recommendation | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 19:07 | yes | 5 | 11 s |
| Direct recommendation | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 09:52 | yes | 5 | 4 s |
| Direct recommendation | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 17:23 | yes | 21 | 63 s |
| Direct recommendation | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 17:48 | yes | 5 | 2 s |
| Direct recommendation | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 17:17 | no | 0 | 26 s |
| Direct recommendation | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 18:22 | yes | 19 | 26 s |
| Direct recommendation | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 11:07 | yes | 24 | 86 s |
| Direct recommendation | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 08:49 | yes | 5 | 37 s |
| Direct recommendation | GPT-6 Luna | gpt-6-luna | 2026-11-01 18:56 | yes | 6 | 17 s |
| Direct recommendation | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 19:21 | yes | 14 | 21 s |
| Paraphrase | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 16:48 | yes | 10 | 8 s |
| Paraphrase | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 13:01 | yes | 6 | 6 s |
| Paraphrase | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 10:02 | no | 0 | 13 s |
| Paraphrase | Perplexity Sonar | sonar | 2026-11-01 17:52 | yes | 10 | 5 s |
| Paraphrase | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 13:49 | yes | 5 | 10 s |
| Paraphrase | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 13:01 | yes | 5 | 5 s |
| Paraphrase | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 17:23 | yes | 19 | 42 s |
| Paraphrase | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 14:23 | yes | 5 | 3 s |
| Paraphrase | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 15:12 | no | 0 | 28 s |
| Paraphrase | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 14:35 | yes | 13 | 56 s |
| Paraphrase | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 19:07 | yes | 20 | 35 s |
| Paraphrase | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 10:58 | no | 0 | 118 s |
| Paraphrase | GPT-6 Luna | gpt-6-luna | 2026-11-01 14:58 | yes | 3 | 16 s |
| Paraphrase | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 10:37 | yes | 13 | 27 s |
| Comparative | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 12:32 | yes | 9 | 8 s |
| Comparative | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 11:27 | yes | 4 | 6 s |
| Comparative | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 09:27 | no | 0 | 18 s |
| Comparative | Perplexity Sonar | sonar | 2026-11-01 18:18 | yes | 10 | 7 s |
| Comparative | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 17:25 | yes | 5 | 14 s |
| Comparative | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 16:28 | yes | 5 | 9 s |
| Comparative | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 09:34 | yes | 20 | 41 s |
| Comparative | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 09:00 | yes | 5 | 7 s |
| Comparative | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 17:46 | no | 0 | 29 s |
| Comparative | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 10:37 | yes | 15 | 25 s |
| Comparative | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 10:02 | yes | 25 | 137 s |
| Comparative | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 10:11 | yes | 5 | 114 s |
| Comparative | GPT-6 Luna | gpt-6-luna | 2026-11-01 16:54 | yes | 6 | 24 s |
| Comparative | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 10:02 | yes | 14 | 23 s |
| Budget constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 11:42 | no | 0 | 5 s |
| Budget constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 19:13 | no | 0 | 3 s |
| Budget constrained | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 09:24 | no | 0 | 15 s |
| Budget constrained | Perplexity Sonar | sonar | 2026-11-01 10:32 | yes | 10 | 3 s |
| Budget constrained | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 16:16 | yes | 5 | 12 s |
| Budget constrained | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 14:33 | yes | 5 | 4 s |
| Budget constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 11:38 | yes | 25 | 30 s |
| Budget constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 17:58 | yes | 5 | 4 s |
| Budget constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 10:36 | yes | 5 | 27 s |
| Budget constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 09:46 | yes | 15 | 66 s |
| Budget constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 09:59 | yes | 25 | 33 s |
| Budget constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 11:01 | yes | 5 | 28 s |
| Budget constrained | GPT-6 Luna | gpt-6-luna | 2026-11-01 12:55 | yes | 3 | 9 s |
| Budget constrained | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 08:35 | yes | 15 | 15 s |
| Scale constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 19:02 | no | 0 | 7 s |
| Scale constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 15:24 | no | 0 | 8 s |
| Scale constrained | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 10:46 | no | 0 | 14 s |
| Scale constrained | Perplexity Sonar | sonar | 2026-11-01 13:43 | yes | 10 | 7 s |
| Scale constrained | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 18:04 | yes | 5 | 11 s |
| Scale constrained | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 10:31 | no | 0 | 6 s |
| Scale constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 17:42 | yes | 23 | 39 s |
| Scale constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 10:27 | yes | 5 | 3 s |
| Scale constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 16:22 | no | 0 | 33 s |
| Scale constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 18:55 | no | 0 | 43 s |
| Scale constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 14:06 | yes | 14 | 39 s |
| Scale constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 17:56 | no | 0 | 28 s |
| Scale constrained | GPT-6 Luna | gpt-6-luna | 2026-11-01 14:22 | yes | 3 | 19 s |
| Scale constrained | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 16:58 | yes | 12 | 23 s |
| Negative framing | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 11:37 | no | 0 | 3 s |
| Negative framing | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 14:55 | yes | 4 | 7 s |
| Negative framing | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 08:54 | no | 0 | 20 s |
| Negative framing | Perplexity Sonar | sonar | 2026-11-01 11:15 | yes | 10 | 4 s |
| Negative framing | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 16:33 | yes | 5 | 53 s |
| Negative framing | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 10:56 | yes | 5 | 4 s |
| Negative framing | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 09:11 | yes | 24 | 22 s |
| Negative framing | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 11:37 | yes | 5 | 3 s |
| Negative framing | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 08:39 | yes | 5 | 51 s |
| Negative framing | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 14:53 | yes | 25 | 30 s |
| Negative framing | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 14:53 | yes | 24 | 117 s |
| Negative framing | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 16:19 | no | 0 | 13 s |
| Negative framing | GPT-6 Luna | gpt-6-luna | 2026-11-01 10:52 | yes | 4 | 11 s |
| Negative framing | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 16:57 | yes | 15 | 31 s |
Normalization in this category
Every judgment call made between the raw labels and the numbers above, listed so it is visible and reversible.
ShowHide
Category-scoped readings
Databricks read as Databricks Mosaic AI
Databricks (MLflow / Mosaic AI) read as Databricks Mosaic AI
Databricks (Mosaic AI / Data Intelligence Platform) read as Databricks Mosaic AI
Databricks (Mosaic AI / MLflow) read as Databricks Mosaic AI
Databricks (Mosaic AI) read as Databricks Mosaic AI
Databricks (Unity Catalog, MLflow) read as Databricks Mosaic AI
Databricks (with Mosaic AI) read as Databricks Mosaic AI
Databricks Community Edition read as Databricks Mosaic AI
Databricks Machine Learning read as Databricks Mosaic AI
Domino read as Domino Data Lab
Google Cloud AI/Vertex AI read as Google Vertex AI
KNIME read as KNIME Analytics Platform
Palantir read as Palantir Foundry
Snowflake read as Snowflake Cortex AI
Snowflake (Cortex/Snowpark) read as Snowflake Cortex AI
Snowflake Cortex read as Snowflake Cortex AI
Snowflake ML read as Snowflake Cortex AI
Spark read as Apache Spark MLlib
Unresolved, counted raw
AWS SageMaker Studio Lab
Altair RapidMiner
Alteryx Analytics
Anthropic's Claude
Apache MLflow
AutoSklearn
Azure Machine Learning, Azure AI Foundry
Civitai
Create ML (Apple)
Databricks Lakehouse Platform
Databricks ML
Databricks MLflow
Databricks Mosaic AI with MLflow
DeepInfra
Fast.ai
GitHub-hosted model repositories
Google Cloud Functions
Google Colaboratory
Google's AutoML
Hugging Face Endpoints
Iguazio
Keras
Kubeflow on-premises / Anthos
Lambdalabs
Microsoft Power BI AI visuals
ModelScope
NVIDIA Dynamo-Triton
Nyckel
OpenAI Platform
OpenCSG
Oracle Cloud Infrastructure (OCI) Data Science
Outerbounds (managed Metaflow)
Ray-based systems
Slack GPT
TPOT
TensorFlow Playground
Triton/TorchServe
Vanna.AI
Discontinued, still offered
No shut-down product was recommended here.