# Feature Requests **URL:** https://community.pinecone.io/c/feature-requests/40.md?page=1 [Latest](https://community.pinecone.io/latest.md) · [Categories](https://community.pinecone.io/categories.md) · [Tags](https://community.pinecone.io/tags.md) **Page:** 2 --- ## [Creation/deletion of collection via API should be deterministic](https://community.pinecone.io/t/creation-deletion-of-collection-via-api-should-be-deterministic/6405)
**Author:** [@n.ninkovic](https://community.pinecone.io/u/n.ninkovic)\ **Replies:** 3\ **Last updated:** [September 30, 2024, 6:56am UTC](https://community.pinecone.io/t/creation-deletion-of-collection-via-api-should-be-deterministic/6405 "2024-09-30T06:56:51Z")
This might be related to the Bulk export feature request, but still writing it here to explain our use case. We have a production index but we also want to allow the development team to test and reproduce issues that oc… --- ## [Assistant: DOC, DOCX and directory or drive support?](https://community.pinecone.io/t/assistant-doc-docx-and-directory-or-drive-support/6638)
**Author:** [@crispin.courtenay](https://community.pinecone.io/u/crispin.courtenay)\ **Replies:** 1\ **Last updated:** [September 23, 2024, 2:39pm UTC](https://community.pinecone.io/t/assistant-doc-docx-and-directory-or-drive-support/6638 "2024-09-23T14:39:19Z")
I have been using assistant for a few days now. It is a brilliant product (most of the time) as it stands. However I have tens of thousands of Word Docs that need to be uploaded, in nested directories or multiple drives … --- ## [Quering with multiple vectors at a time](https://community.pinecone.io/t/quering-with-multiple-vectors-at-a-time/6566)
**Author:** [@gayani.parameswaran](https://community.pinecone.io/u/gayani.parameswaran)\ **Replies:** 1\ **Last updated:** [September 13, 2024, 7:11pm UTC](https://community.pinecone.io/t/quering-with-multiple-vectors-at-a-time/6566 "2024-09-13T19:11:03Z")
Is it possible to query multiple vectors in a single request by passing several vectors at once? I need to retrieve information for five data points simultaneously, and performing sequential retrieval increases latency. … --- ## [Limit API Key access by Index](https://community.pinecone.io/t/limit-api-key-access-by-index/6570)
**Author:** [@andy-tmpt-me](https://community.pinecone.io/u/andy-tmpt-me)\ **Replies:** 2\ **Last updated:** [September 13, 2024, 4:23pm UTC](https://community.pinecone.io/t/limit-api-key-access-by-index/6570 "2024-09-13T16:23:44Z")
When creating an API key, I would like to select which indexes can be accessed by the API key (and ideally whether the key is read only or read write). In order to accomplish a use case where my Dev team cannot inadverte… --- ## [Paginate search results](https://community.pinecone.io/t/paginate-search-results/540)
**Author:** [@moa](https://community.pinecone.io/u/moa)\ **Replies:** 3\ **Last updated:** [September 12, 2024, 1:10pm UTC](https://community.pinecone.io/t/paginate-search-results/540 "2024-09-12T13:10:09Z")
I need to display the results in pages of N documents and show the user how many total results are available Is there a way to get use skip like other databases like mongo? Is there a way to get histogram of scores? eg … --- ## [Still waiting for a drag-and-drop embedding app or service](https://community.pinecone.io/t/still-waiting-for-a-drag-and-drop-embedding-app-or-service/6393)
**Author:** [@gaoptimize](https://community.pinecone.io/u/gaoptimize)\ **Replies:** 6\ **Last updated:** [August 23, 2024, 2:06pm UTC](https://community.pinecone.io/t/still-waiting-for-a-drag-and-drop-embedding-app-or-service/6393 "2024-08-23T14:06:39Z")
I’d be a significant Pinecone user if there were a desk top Windows program or web service that did batch document (MS Office and pdf) encoding to Pinecone index. This is a significant opportunity for an enterprising de… --- ## [Update documentation/product roadmap for metadata filters](https://community.pinecone.io/t/update-documentation-product-roadmap-for-metadata-filters/6324)
**Author:** [@softservesoftware](https://community.pinecone.io/u/softservesoftware)\ **Replies:** 1\ **Last updated:** [August 20, 2024, 2:09pm UTC](https://community.pinecone.io/t/update-documentation-product-roadmap-for-metadata-filters/6324 "2024-08-20T14:09:41Z")
Since the pinecone metadata filters are based on MongoDB query/projection operators (ref: pinecone docs) it is assumed that there should be full support for all operators that exist in the mongo documentation. The pinec… --- ## [Retrieve all vector IDs](https://community.pinecone.io/t/retrieve-all-vector-ids/3421)
**Author:** [@Cory\_Pinecone](https://community.pinecone.io/u/Cory_Pinecone)\ **Replies:** 10\ **Last updated:** [August 16, 2024, 8:45am UTC](https://community.pinecone.io/t/retrieve-all-vector-ids/3421 "2024-08-16T08:45:38Z")
An endpoint to retrieve all of the vector IDs stored in the index without supplying a query vector first. --- ## [setMetadata deprecated in 3.0.0 :(](https://community.pinecone.io/t/setmetadata-deprecated-in-3-0-0/6258)
**Author:** [@david.oates](https://community.pinecone.io/u/david.oates)\ **Replies:** 1\ **Last updated:** [August 16, 2024, 8:41am UTC](https://community.pinecone.io/t/setmetadata-deprecated-in-3-0-0/6258 "2024-08-16T08:41:47Z")
According to Pinecone support, the ability to overwrite the entire metadata object without a full upsert has been removed in 3.0.0. That means if you want to remove a field from metadata (instead of just change it) you … --- ## [Please take a look at OpenAI's "Assistants" UI and implement something like it](https://community.pinecone.io/t/please-take-a-look-at-openais-assistants-ui-and-implement-something-like-it/4122)
**Author:** [@gaoptimize](https://community.pinecone.io/u/gaoptimize)\ **Replies:** 1\ **Last updated:** [August 16, 2024, 8:40am UTC](https://community.pinecone.io/t/please-take-a-look-at-openais-assistants-ui-and-implement-something-like-it/4122 "2024-08-16T08:40:39Z")
I’ve spent too many hours trying to get to the point I could upsert 2 or 3 hundred 20KBye to 2MByte text files to the Pinecone index/environment I created. The good news is Canopy – no-rag is working in Python. The bad… --- ## [Sync between multiple indexes](https://community.pinecone.io/t/sync-between-multiple-indexes/6289)
**Author:** [@arnab](https://community.pinecone.io/u/arnab)\ **Replies:** 1\ **Last updated:** [August 5, 2024, 4:12pm UTC](https://community.pinecone.io/t/sync-between-multiple-indexes/6289 "2024-08-05T16:12:40Z")
What’s the best way to sync between the dev and prod index? Our regular RDS database sync happens once a week or two. And we are looking for the same in Pinecone VectorDB too. Is there an easier, cost-effective way to s… --- ## [Concurrency model for updating data](https://community.pinecone.io/t/concurrency-model-for-updating-data/5859)
**Author:** [@eds228](https://community.pinecone.io/u/eds228)\ **Replies:** 1\ **Last updated:** [June 17, 2024, 12:33pm UTC](https://community.pinecone.io/t/concurrency-model-for-updating-data/5859 "2024-06-17T12:33:38Z")
I want to store information in the rows of the metadata. For example metadata= metadata={"count": 5} I want concurrent processes to be able to increment the count. Does pinecone expose any consistency mechanism to implem… --- ## [Selective indexing at namespace level](https://community.pinecone.io/t/selective-indexing-at-namespace-level/5755)
**Author:** [@infra](https://community.pinecone.io/u/infra)\ **Replies:** 3\ **Last updated:** [June 10, 2024, 8:34pm UTC](https://community.pinecone.io/t/selective-indexing-at-namespace-level/5755 "2024-06-10T20:34:30Z")
For pod-based indexes, Pinecone indexes all metadata by default. Use selective metadata indexing to prevent this Request 1: would be highlight this somewhere in starter notebooks, as many aren’t aware of this issue unti… --- ## [Similarity Search with score cutoff](https://community.pinecone.io/t/similarity-search-with-score-cutoff/5822)
**Author:** [@johnkongtcheu](https://community.pinecone.io/u/johnkongtcheu)\ **Replies:** 5\ **Last updated:** [June 8, 2024, 4:39am UTC](https://community.pinecone.io/t/similarity-search-with-score-cutoff/5822 "2024-06-08T04:39:21Z")
Hello, Pinecone community, I’m writing to ask if it’s possible to have a vector similarity search with thresholding instead of just top K. Similar to what Weaviate has(Vector similarity search | Weaviate - Vector Databa… --- ## [Can I add multiple embeddings into one record and run similarity searches on multiple embeddings?](https://community.pinecone.io/t/can-i-add-multiple-embeddings-into-one-record-and-run-similarity-searches-on-multiple-embeddings/5641)
**Author:** [@jason.kang](https://community.pinecone.io/u/jason.kang)\ **Replies:** 4\ **Last updated:** [May 23, 2024, 2:41pm UTC](https://community.pinecone.io/t/can-i-add-multiple-embeddings-into-one-record-and-run-similarity-searches-on-multiple-embeddings/5641 "2024-05-23T14:41:37Z")
Same as the title, I have been using Azure AI Search as a vector database of which columns are content and summary. and I have been running similarity searches on both the content field and summary field at the same time… --- ## ["string includes" metadata filter](https://community.pinecone.io/t/string-includes-metadata-filter/5219)
**Author:** [@vincent.coraldean](https://community.pinecone.io/u/vincent.coraldean)\ **Replies:** 1\ **Last updated:** [May 6, 2024, 6:29pm UTC](https://community.pinecone.io/t/string-includes-metadata-filter/5219 "2024-05-06T18:29:04Z")
I’m not sure whether this would be useful to others, but it would certainly be useful for our use-cases on our current RAG-driven application project. Something like $incl and $excl would be great for filtering metadata … --- ## [Looking for change data capture (CDC) notification or trigger option](https://community.pinecone.io/t/looking-for-change-data-capture-cdc-notification-or-trigger-option/5300)
**Author:** [@santhosh.sundarasamy](https://community.pinecone.io/u/santhosh.sundarasamy)\ **Replies:** 0\ **Last updated:** [April 22, 2024, 3:33pm UTC](https://community.pinecone.io/t/looking-for-change-data-capture-cdc-notification-or-trigger-option/5300 "2024-04-22T15:33:26Z")
Looking for CDC notification or trigger - like when new vector is added registered webhook is called or some form of notification is sent… this is critical for passive apps to give seamless flow of integration. --- ## [Get number of unique values of a metadata field](https://community.pinecone.io/t/get-number-of-unique-values-of-a-metadata-field/5234)
**Author:** [@matined](https://community.pinecone.io/u/matined)\ **Replies:** 1\ **Last updated:** [April 19, 2024, 9:44pm UTC](https://community.pinecone.io/t/get-number-of-unique-values-of-a-metadata-field/5234 "2024-04-19T21:44:08Z")
Hello, I’d like to get the number of unique vectors in my Pinecone index but with grouping by a metadata field. In the other words, get the number of unique values of a metadata field. Is it possible in Pinecone? --- ## ["Starts with" metadata filter](https://community.pinecone.io/t/starts-with-metadata-filter/3579)
**Author:** [@Jasper](https://community.pinecone.io/u/Jasper)\ **Replies:** 5\ **Last updated:** [April 1, 2024, 11:39pm UTC](https://community.pinecone.io/t/starts-with-metadata-filter/3579 "2024-04-01T23:39:09Z")
A full regex metadata filter would be nice, but a simpler “starts with” would be very nice for a start. --- ## [Pinecone and Docker](https://community.pinecone.io/t/pinecone-and-docker/4877)
**Author:** [@joeykees](https://community.pinecone.io/u/joeykees)\ **Replies:** 2\ **Last updated:** [March 25, 2024, 1:16pm UTC](https://community.pinecone.io/t/pinecone-and-docker/4877 "2024-03-25T13:16:29Z")
I was wondering if there are any plans for docker support? Or have misread something? --- ## [Support for $exists operator while metadata filtering](https://community.pinecone.io/t/support-for-exists-operator-while-metadata-filtering/3680)
**Author:** [@shubham.pawar](https://community.pinecone.io/u/shubham.pawar)\ **Replies:** 7\ **Last updated:** [March 18, 2024, 7:03pm UTC](https://community.pinecone.io/t/support-for-exists-operator-while-metadata-filtering/3680 "2024-03-18T19:03:08Z")
MongoDB supports an element operator called $exists which matches documents that have the specified field. It would offset for the lack of support for NULL metadata values. MongoDB docs: https://www.mongodb.com/docs/man… --- ## [Web or desktop app to optimize and perform chunking](https://community.pinecone.io/t/web-or-desktop-app-to-optimize-and-perform-chunking/4886)
**Author:** [@gaoptimize](https://community.pinecone.io/u/gaoptimize)\ **Replies:** 0\ **Last updated:** [March 16, 2024, 4:14pm UTC](https://community.pinecone.io/t/web-or-desktop-app-to-optimize-and-perform-chunking/4886 "2024-03-16T16:14:46Z")
I suspect that the experts at Pinecone know enough about RAG performance issues (both quality and speed) and chunking to be able to write a program or develop a web interface to optimize chunking of data into Pinecone, s… --- ## [Support collection for serverless index](https://community.pinecone.io/t/support-collection-for-serverless-index/4863)
**Author:** [@dishen](https://community.pinecone.io/u/dishen)\ **Replies:** 1\ **Last updated:** [March 16, 2024, 2:19pm UTC](https://community.pinecone.io/t/support-collection-for-serverless-index/4863 "2024-03-16T14:19:19Z")
Hello, will collections be supported for serverless indexes in the future? --- ## [Listing records with metadata field filtering](https://community.pinecone.io/t/listing-records-with-metadata-field-filtering/4874)
**Author:** [@AusafMo](https://community.pinecone.io/u/AusafMo)\ **Replies:** 0\ **Last updated:** [March 15, 2024, 5:54am UTC](https://community.pinecone.io/t/listing-records-with-metadata-field-filtering/4874 "2024-03-15T05:54:49Z")
As of now, there is no way to directly filter all the records using metadata fields while using list(). Altough there’s a workaround, for more context see Having the ability to add filters while listing the records usi… --- ## [Support for RAG Fusion](https://community.pinecone.io/t/support-for-rag-fusion/4031)
**Author:** [@PeterCotroneo](https://community.pinecone.io/u/PeterCotroneo)\ **Replies:** 1\ **Last updated:** [March 13, 2024, 10:28am UTC](https://community.pinecone.io/t/support-for-rag-fusion/4031 "2024-03-13T10:28:35Z")
Hi, I’d like to request support for RAG Fusion. Essentially, it addresses a limitation in RAG by generating multiple queries and reranking the results (using Reciprocal Rank Fusion and custom vector score weighting). H… --- ## [Query by vector across Namespaces](https://community.pinecone.io/t/query-by-vector-across-namespaces/4424)
**Author:** [@nsartor](https://community.pinecone.io/u/nsartor)\ **Replies:** 0\ **Last updated:** [February 5, 2024, 3:25pm UTC](https://community.pinecone.io/t/query-by-vector-across-namespaces/4424 "2024-02-05T15:25:00Z")
Hi, We use Pinecone to store embeddings of different text “types”, which I organized in namespaces. My major use case is to query for an id from namespace A and want results from namespace B. Is this something you hav… --- ## [429 Error while calling /vectors/list api endpoint](https://community.pinecone.io/t/429-error-while-calling-vectors-list-api-endpoint/4382)
**Author:** [@anni1236012](https://community.pinecone.io/u/anni1236012)\ **Replies:** 0\ **Last updated:** [February 1, 2024, 5:43pm UTC](https://community.pinecone.io/t/429-error-while-calling-vectors-list-api-endpoint/4382 "2024-02-01T17:43:52Z")
I recently started using Pinecone Serverless and encountered an issue with deleting vectors. I’ve noticed that I’m consistently hitting a rate limit with a 429 error, even when I’m making only around 20 asynchronous API … --- ## [Request a speaker at your developer event](https://community.pinecone.io/t/request-a-speaker-at-your-developer-event/4094)
**Author:** [@jocelyn](https://community.pinecone.io/u/jocelyn)\ **Replies:** 0\ **Last updated:** [January 16, 2024, 5:49am UTC](https://community.pinecone.io/t/request-a-speaker-at-your-developer-event/4094 "2024-01-16T05:49:01Z")
Do you have an upcoming developer-focused meetup or webinar focused on Pinecone, LLM or RAG? Our developer relations team regularly gives live talks, webinars, podcasts. We also write articles and give demos. Focused on… --- ## [.pdf to text. Best options suggestions please](https://community.pinecone.io/t/pdf-to-text-best-options-suggestions-please/4051)
**Author:** [@gaoptimize](https://community.pinecone.io/u/gaoptimize)\ **Replies:** 0\ **Last updated:** [January 11, 2024, 12:48am UTC](https://community.pinecone.io/t/pdf-to-text-best-options-suggestions-please/4051 "2024-01-11T00:48:30Z")
I know Audrey said on 9 Jan that .pdf upload is coming. The issue is VERY important to me, and so much so that it could drive me to Perplexity Pro (but it has a persistence issue unfortunately). In the meantime, I’ve t… --- ## [Canopy-specific category?](https://community.pinecone.io/t/canopy-specific-category/3917)
**Author:** [@gaoptimize](https://community.pinecone.io/u/gaoptimize)\ **Replies:** 2\ **Last updated:** [December 28, 2023, 4:22pm UTC](https://community.pinecone.io/t/canopy-specific-category/3917 "2023-12-28T16:22:02Z")
I don’t want to bother the expert Pinecone users here with Canopy-related qustions, so I was hoping it could have its own category in this forum. I’m having trouble with where pip is putting Canopy. I did a System inst… [Previous page](https://community.pinecone.io/c/feature-requests/40.md) [Next page](https://community.pinecone.io/c/feature-requests/40.md?page=2)