Vector databases for enterprise buyers
Asked as “vector database”, and as “vector search database for AI applications”, on behalf of an enterprise B2B company. 43 first choices recorded across the direct, paraphrase, budget and scale prompts, twelve models each.
Standing · first-choice share
53%
Clear leader
53Pinecone19Qdrant12pgvector16others
53% of first choices, clear leader.
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.
The standing
Share is the count of first choices across the direct, paraphrase, budget and scale prompts, over all twelve models, for an enterprise B2B company. Ordered by share.
ProductFirst-choice shareNegative rateLabelsQuadrant
Show the five products at 0%, ordered by negative rate
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.
Head to headHead-to-head pages are published for mid-market buyers, so these open on mid-market figures, not enterprise.
All fifteen head-to-head pages: the top six products, each against each
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 →
01
02
03
04
05
06
07
08E09
R10
11Accepted challengerEndorsed leader
0%First-choice share → · lines at 30% share and 25% negative70%
Key
01Pinecone53%
Pinecone53%02Qdrant19%
Qdrant19%03pgvector12%
pgvector12%04Weaviate9%
Weaviate9%05Zilliz Cloud5%
Zilliz Cloud5%06Milvus / Zilliz Cloud2%
Milvus / Zilliz Cloud2%07Milvus0%
Milvus0%08Azure AI Search0%
Azure AI Search0%09Elasticsearch0%
10Redis Vector Search0%
11Chroma0%
Chroma0%What they warned about
Six of twelve models held their first choice under the paraphrase. Gemini 3.5 Flash, Perplexity Sonar, Qwen 3.7 Flash, Kimi K2, GLM 4.7 FlashX and MiniMax M2.5 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.
Pinecone28%
18 of 64 labels negative · 11 of 12 models · 4 hard negative
“Avoid "Serverless" plans (like Pinecone Serverless or standard AWS Opensearch Vector Engine) for thousands of users.” Qwen 3.7 Flash, budget prompt
pgvector31%
17 of 54 labels negative · 8 of 12 models
“Postgres + pgvector if you already standardize on Postgres and want to inherit existing controls, though you may trade off scale and specialized search features” GPT-5.4 mini, scale prompt
Chroma67%
10 of 15 labels negative · 7 of 12 models · 6 hard negative
“Many "developer‑friendly" tools (e.g., Chroma) are excellent for prototypes but lack enterprise features ... avoid them for mission‑critical RAG pipelines” GLM 4.7 FlashX, negative prompt
Milvus20%
8 of 41 labels negative · 7 of 12 models · 1 hard negative
“avoid highly complex C++ or Python-based microservices architectures like Milvus or Zilliz (when self-hosted) unless you have a dedicated Site Reliability Engineering (SRE) team” Qwen 3.7 Flash, negative prompt
What they cite
Twelve of the twelve models returned a source list in this edition: four return the pages they cited and eight the search results they consulted.
Sites the answers cite
67 of 72 answers in this category came back with a source list, from 12 of 12 models: citations where the model returns them, or the search results it consulted. 928 links across 281 sites, every framing counted. Ranked by the number of answers carrying the site or page.
vendor site · Firecrawl35 answers · 36 citations · 11 models
vendor site · Atlan32 answers · 32 citations · 9 models
31 answers · 31 citations · 10 models
27 answers · 28 citations · 9 models
25 answers · 32 citations · 10 models
21 answers · 28 citations · 8 models
21 answers · 21 citations · 9 models
20 answers · 20 citations · 8 models
19 answers · 20 citations · 7 models
vendor site · Actian18 answers · 21 citations · 10 models
vendor site · G218 answers · 18 citations · 8 models
vendor site · Pinecone17 answers · 26 citations · 8 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.
35 answers · 11 models
32 answers · 9 models
31 answers · 10 models
27 answers · 9 models
21 answers · 9 models
19 answers · 7 models
19 answers · 8 models
18 answers · 8 models
14 answers · 8 models
10 answers · 8 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 vector database”, “vector database”, “vector databases”. 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 | Google ads, last 30 days | Best Google position |
|---|---|---|---|---|---|---|---|
| Pinecone pinecone.io | 53% | 188 | 60,500 | 3,929 | 90,633 | 14 | 4 |
| Qdrant qdrant.tech | 19% | 186 | 6,600 | 180 | 8,064 | 24 | |
| pgvector pgvector.app | 12% | 172 | 5,400 | 119 | |||
| Weaviate weaviate.io | 9% | 158 | 2,900 | 54 | 5,054 | ||
| Zilliz Cloud zilliz.com | 5% | 26 | 140 | 2 | 11,801 | 24 | 7 |
| Milvus milvus.io Milvus / Zilliz Cloud 2%, Milvus 0% | 2% | 96 | 3,600 | 169 | 52,360 | ||
| Azure azure.com Azure AI Search 0% | 0% | 22 | 301,000 | 47,904 | |||
| Elasticsearch | 0% | 12,100 | 2,798 | ||||
| Vespa vespa.ai | 0% | 7 | 135,000 | 12,778 | 1,831 | 12 | |
| Redis Vector Search redis.io | 0% | 17 | 112,424 | 18 |
Google's top ten for the three searches hold 16 sites; 4 of them are among the sites the models cited here (encore.dev, g2.com, medium.com, pinecone.io). The first result for “best vector database” is zenml.io.
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 vector database for an enterprise B2B company?”Paraphrase“Which vector search database for AI applications would you recommend to a large B2B company with thousands of employees?”Comparative“What are the top enterprise-grade vector databases and how do they differ?”Budget-constrained“What is the best vector database for a large company that needs predictable total cost across thousands of users?”Scale-constrained“We are a 5,000 person company with SSO, SOC 2 and procurement review requirements evaluating a vector database. What should we look at?”Negative“Which vector databases should a large enterprise avoid or be cautious about?”
Claude Haiku 4.5PineconePineconeHeldPinecone, QdrantQdrant, Weaviateagainst: Pineconeno first choiceagainst: FAISS, Milvus
Four alternatives
Milvus, Weaviate, Zilliz Cloud, pgvectorThree alternatives
Milvus, Weaviate, pgvectorTwo alternatives
Milvus, pgvectorOne alternative
pgvectorGPT-5.4 miniPineconePineconeHeldPineconeagainst: Elasticsearch / OpenSearch, pgvectorPineconeno first choiceagainst: pgvectornothing named
Two alternatives
Milvus, WeaviateThree alternatives
Amazon OpenSearch / Aurora / DynamoDB / Neptune, Google Cloud Vector Search / Agent Retrieval, QdrantFour alternatives
Azure AI Search, Milvus, MongoDB Atlas Vector Search, WeaviateThree alternatives
Milvus, Qdrant, WeaviateGemini 3.5 FlashpgvectorPineconeChangedpgvectoragainst: Elasticsearch / OpenSearch, Redis Vector SearchQdrantagainst: Pinecone, Weaviatepgvectoragainst: Chroma, FAISS, HNSWlib, LanceDB, Milvus, MongoDB Atlas Vector Search, Pinecone, ScaNN, pgvector
Four alternatives
Milvus, Pinecone, Qdrant, WeaviateFour alternatives
Milvus, Qdrant, Weaviate, pgvectorThree alternatives
Milvus, Pinecone, WeaviateTwo alternatives
Milvus / Zilliz Cloud, `pgvector` with `pgvectorscale`Seven alternatives
Elasticsearch / OpenSearch, Milvus, MongoDB Atlas Vector Search, Pinecone, Qdrant, Weaviate, Zilliz CloudPerplexity SonarPinecone, pgvectorPineconeChangedagainst: pgvectorPineconeagainst: pgvectorQdrantno first choiceagainst: Chroma, FAISS, Pinecone, pgvector
Three alternatives
Milvus, Qdrant, WeaviateFour alternatives
Azure AI Search, Elasticsearch, Milvus, WeaviateThree alternatives
Milvus / Zilliz Cloud, Qdrant, WeaviateTwo alternatives
Mixpeek Vector Store, pgvectorGrok 4.20Pineconeagainst: Oracle, pgvectorPineconeHeldPinecone, Zilliz Cloudagainst: Chroma, Elasticsearch, LanceDBQdrantagainst: Pinecone, Weaviate, Zilliz CloudPineconeagainst: Milvus, pgvectoragainst: Chroma, Milvus, Pinecone, Weaviate, pgvector
Two alternatives
Weaviate, Zilliz CloudFour alternatives
Milvus / Zilliz Cloud, Qdrant, Weaviate, pgvectorFour alternatives
Milvus, Qdrant, Weaviate, pgvectorTwo alternatives
Milvus, WeaviateFour alternatives
Chroma, Qdrant, Weaviate, Zilliz CloudMistral SmallPinecone, Weaviateagainst: MilvusPinecone, WeaviateHeldno first choiceMilvus / Zilliz Cloudagainst: Weaviateno first choiceagainst: FAISS, Pinecone
One alternative
pgvectorTwo alternatives
Azure AI Search, MilvusOne alternative
PineconeDeepSeek V4 FlashPineconePineconeHeldMilvus / Zilliz Cloud, Pinecone, Qdrant, Weaviateagainst: Chroma, LanceDBQdrant, Weaviateagainst: Pineconeno first choiceagainst: Chroma, Milvus, Pinecone, Weaviate
Four alternatives
Milvus / Zilliz Cloud, Qdrant, Weaviate, pgvectorSix alternatives
Azure AI Search, Milvus / Zilliz Cloud, Qdrant, Vertex AI Vector Search, Weaviate, pgvectorTwo alternatives
Redis Vector Search, pgvectorOne alternative
MilvusLlama 4 Maverickno first choiceno first choiceHeldno first choiceno first choiceagainst: Pinecone, pgvectorPineconeagainst: Chroma, Pinecone, pgvector
Two alternatives
Qdrant, WeaviateFour alternatives
Azure AI Search, Azure OpenAI Service, Milvus, WeaviateQwen 3.7 FlashpgvectorPineconeChangedno first choicePineconeagainst: OpenSearch, Pineconeno first choiceagainst: FAISS, Milvus, Numpy, Pinecone, Zilliz Cloud
Four alternatives
Elasticsearch, Milvus, Pinecone, WeaviateThree alternatives
Milvus, Weaviate, pgvectorSix alternatives
Elasticsearch, Milvus, Pinecone, Qdrant, Weaviate, Zilliz CloudThree alternatives
Milvus, Qdrant, pgvectorKimi K2Pinecone, Zilliz CloudPineconeChangedagainst: pgvectorPineconeagainst: Elasticsearch, pgvectorQdrantagainst: PineconePineconeagainst: pgvectoragainst: Annoy, Chroma, FAISS, Pinecone, Qdrant, Weaviate
Two alternatives
Weaviate, pgvectorTwo alternatives
Weaviate, Zilliz CloudThree alternatives
Milvus / Zilliz Cloud, Qdrant, WeaviateOne alternative
MilvusFour alternatives
Milvus, Qdrant, Weaviate, Zilliz CloudGLM 4.7 FlashXPineconeQdrantChangedagainst: pgvectorPineconeagainst: ChromaZilliz Cloudagainst: Pineconeno first choiceagainst: Chroma, Elasticsearch, MongoDB Atlas Vector Search, Redis Vector Search, pgvector
Four alternatives
Milvus / Zilliz Cloud, Qdrant, Weaviate, pgvectorThree alternatives
Milvus / Zilliz Cloud, Pinecone, WeaviateEight alternatives
Elasticsearch, Milvus / Zilliz Cloud, MongoDB Atlas Vector Search, Qdrant, Redis Vector Search, Vespa, Weaviate, pgvectorTwo alternatives
Qdrant, pgvectorMiniMax M2.5pgvectoragainst: PineconePineconeChangedPineconeQdrantagainst: Upstash Vectorno first choiceagainst: Milvus, Pinecone, Weaviate, pgvector
Two alternatives
Milvus, WeaviateThree alternatives
Azure AI Search, Weaviate, pgvectorSeven alternatives
Elasticsearch, Milvus / Zilliz Cloud, Qdrant, Redis Vector Search, Vespa, Weaviate, pgvectorTwo alternatives
Pinecone, Zilliz CloudBold 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.
Seventy-two 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 | 2026-09-17 10:38 | yes | 16 | 11 s |
| Direct recommendation | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-09-17 11:44 | yes | 5 | 7 s |
| Direct recommendation | Gemini 3.5 Flash | gemini-3.5-flash | 2026-09-17 11:00 | yes | 11 | 26 s |
| Direct recommendation | Perplexity Sonar | sonar | 2026-09-17 09:27 | yes | 20 | 6 s |
| Direct recommendation | Grok 4.20 | spacexai/grok-4.1-fast-non-reasoning via vertex (asked for grok-4.20-non-reasoning; the API answered as this version) | 2026-09-17 13:22 | yes | 38 | 21 s |
| Direct recommendation | Mistral Small | mistral/mistral-small via mistral | 2026-09-17 14:02 | yes | 5 | 5 s |
| Direct recommendation | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via fireworks | 2026-09-17 12:49 | yes | 10 | 15 s |
| Direct recommendation | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-09-17 13:32 | yes | 5 | 3 s |
| Direct recommendation | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-09-17 10:16 | no | 0 | 43 s |
| Direct recommendation | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-09-17 13:35 | yes | 15 | 33 s |
| Direct recommendation | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-09-17 10:09 | yes | 15 | 45 s |
| Direct recommendation | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-09-17 12:13 | yes | 9 | 22 s |
| Paraphrase | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-09-17 10:41 | yes | 9 | 8 s |
| Paraphrase | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-09-17 09:37 | yes | 4 | 5 s |
| Paraphrase | Gemini 3.5 Flash | gemini-3.5-flash | 2026-09-17 13:56 | yes | 8 | 22 s |
| Paraphrase | Perplexity Sonar | sonar | 2026-09-17 10:45 | yes | 20 | 5 s |
| Paraphrase | Grok 4.20 | spacexai/grok-4.1-fast-non-reasoning via vertex (asked for grok-4.20-non-reasoning; the API answered as this version) | 2026-09-17 11:26 | yes | 37 | 16 s |
| Paraphrase | Mistral Small | mistral/mistral-small via mistral | 2026-09-17 09:58 | yes | 5 | 6 s |
| Paraphrase | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via fireworks | 2026-09-17 13:34 | yes | 19 | 23 s |
| Paraphrase | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-09-17 11:35 | yes | 5 | 3 s |
| Paraphrase | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-09-17 12:21 | no | 0 | 40 s |
| Paraphrase | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-09-17 12:16 | yes | 19 | 84 s |
| Paraphrase | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-09-17 13:10 | yes | 15 | 24 s |
| Paraphrase | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-09-17 14:02 | yes | 9 | 41 s |
| Comparative | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-09-17 10:17 | yes | 9 | 9 s |
| Comparative | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-09-17 11:59 | yes | 7 | 12 s |
| Comparative | Gemini 3.5 Flash | gemini-3.5-flash | 2026-09-17 12:32 | yes | 23 | 31 s |
| Comparative | Perplexity Sonar | sonar | 2026-09-17 12:15 | yes | 20 | 12 s |
| Comparative | Grok 4.20 | spacexai/grok-4.1-fast-non-reasoning via vertex (asked for grok-4.20-non-reasoning; the API answered as this version) | 2026-09-17 12:27 | yes | 21 | 14 s |
| Comparative | Mistral Small | mistral/mistral-small via mistral | 2026-09-17 11:59 | yes | 5 | 8 s |
| Comparative | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via fireworks | 2026-09-17 09:54 | yes | 18 | 23 s |
| Comparative | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-09-17 11:40 | yes | 5 | 3 s |
| Comparative | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-09-17 13:05 | no | 0 | 40 s |
| Comparative | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-09-17 13:32 | yes | 15 | 36 s |
| Comparative | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-09-17 11:00 | yes | 13 | 82 s |
| Comparative | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-09-17 10:38 | yes | 5 | 19 s |
| Budget constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-09-17 12:00 | yes | 14 | 11 s |
| Budget constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-09-17 09:37 | yes | 3 | 5 s |
| Budget constrained | Gemini 3.5 Flash | gemini-3.5-flash | 2026-09-17 12:11 | yes | 13 | 26 s |
| Budget constrained | Perplexity Sonar | sonar | 2026-09-17 13:41 | yes | 17 | 9 s |
| Budget constrained | Grok 4.20 | spacexai/grok-4.1-fast-non-reasoning via vertex (asked for grok-4.20-non-reasoning; the API answered as this version) | 2026-09-17 14:02 | yes | 46 | 20 s |
| Budget constrained | Mistral Small | mistral/mistral-small via mistral | 2026-09-17 10:49 | yes | 5 | 8 s |
| Budget constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via fireworks | 2026-09-17 10:08 | yes | 16 | 19 s |
| Budget constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-09-17 11:49 | yes | 5 | 4 s |
| Budget constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-09-17 11:37 | yes | 5 | 35 s |
| Budget constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-09-17 11:03 | yes | 14 | 30 s |
| Budget constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-09-17 11:50 | yes | 27 | 77 s |
| Budget constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-09-17 12:07 | yes | 15 | 67 s |
| Scale constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-09-17 13:16 | yes | 27 | 11 s |
| Scale constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-09-17 14:12 | yes | 5 | 14 s |
| Scale constrained | Gemini 3.5 Flash | gemini-3.5-flash | 2026-09-17 10:38 | no | 0 | 22 s |
| Scale constrained | Perplexity Sonar | sonar | 2026-09-17 09:39 | yes | 20 | 17 s |
| Scale constrained | Grok 4.20 | spacexai/grok-4.1-fast-non-reasoning via vertex (asked for grok-4.20-non-reasoning; the API answered as this version) | 2026-09-17 10:26 | yes | 18 | 12 s |
| Scale constrained | Mistral Small | mistral/mistral-small via mistral | 2026-09-17 14:05 | no | 0 | 7 s |
| Scale constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via fireworks | 2026-09-17 12:06 | yes | 26 | 39 s |
| Scale constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-09-17 13:10 | yes | 5 | 5 s |
| Scale constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-09-17 12:21 | yes | 5 | 37 s |
| Scale constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-09-17 13:41 | yes | 18 | 85 s |
| Scale constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-09-17 10:44 | yes | 14 | 32 s |
| Scale constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-09-17 13:31 | yes | 5 | 32 s |
| Negative framing | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-09-17 11:10 | yes | 9 | 8 s |
| Negative framing | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-09-17 10:36 | yes | 6 | 9 s |
| Negative framing | Gemini 3.5 Flash | gemini-3.5-flash | 2026-09-17 11:05 | yes | 22 | 30 s |
| Negative framing | Perplexity Sonar | sonar | 2026-09-17 11:32 | yes | 20 | 8 s |
| Negative framing | Grok 4.20 | spacexai/grok-4.1-fast-non-reasoning via vertex (asked for grok-4.20-non-reasoning; the API answered as this version) | 2026-09-17 10:48 | yes | 20 | 19 s |
| Negative framing | Mistral Small | mistral/mistral-small via mistral | 2026-09-17 13:15 | yes | 5 | 8 s |
| Negative framing | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via fireworks | 2026-09-17 12:46 | yes | 16 | 30 s |
| Negative framing | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-09-17 10:16 | yes | 5 | 4 s |
| Negative framing | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-09-17 10:28 | yes | 5 | 47 s |
| Negative framing | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-09-17 14:01 | yes | 29 | 44 s |
| Negative framing | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-09-17 11:56 | yes | 14 | 72 s |
| Negative framing | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-09-17 11:52 | yes | 14 | 38 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
MongoDB read as MongoDB Atlas Vector Search
MongoDB Atlas read as MongoDB Atlas Vector Search
PostgreSQL (+ pgvector) read as pgvector
PostgreSQL (via pgvector) read as pgvector
Redis read as Redis Vector Search
Redis (Enterprise) read as Redis Vector Search
Redis (Redis VL) read as Redis Vector Search
Redis (with RedisSearch or RedisStack) read as Redis Vector Search
Redis Enterprise read as Redis Vector Search
Redis Stack read as Redis Vector Search
Redis Vector read as Redis Vector Search
Supabase read as Supabase pgvector
Unresolved, counted raw
Amazon OpenSearch / Aurora / DynamoDB / Neptune
Databricks Vector Search
Google Cloud Vector Search / Agent Retrieval
Mixpeek Vector Store (MVS)
Numpy
TiDB Cloud Serverless
`pgvector` with `pgvectorscale`
Discontinued, still offered
No shut-down product was recommended here.