# Any plans to support new voyage embeddings or jina v2 reranker? **URL:** \ **Category:** General\ **Created:** [July 14, 2024, 1:09am UTC](https://forum.weaviate.io/t/any-plans-to-support-new-voyage-embeddings-or-jina-v2-reranker/3038 "2024-07-14T01:09:01Z")\ **Posts on this page:** 4\ **Page:** 1
**Author:** ![Baptiste\_Cumin1](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.weaviate.io/baptiste_cumin1/32/1943_2.png) [@Baptiste\_Cumin1](https://forum.weaviate.io/u/Baptiste_Cumin1)\ **Post date:** [July 14, 2024, 1:09am UTC](https://forum.weaviate.io/t/any-plans-to-support-new-voyage-embeddings-or-jina-v2-reranker/3038/1 "2024-07-14T01:09:01Z")
What it says on the tin 🙂 Love these models but the lack of support means I’m tempted to go around weaviate which defeats a part of the promise of the product for me! 3 models on my wishlist: - Voyage multilingual v2 embeddings [Embeddings](https://docs.voyageai.com/docs/embeddings) - Jina clip embeddings - Jina reranker v2 [Reranker API](https://jina.ai/reranker/) ---
**Author:** ![DudaNogueira](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.weaviate.io/dudanogueira/32/7846_2.png) [@DudaNogueira](https://forum.weaviate.io/u/DudaNogueira)\ **Post date:** [July 15, 2024, 8:49pm UTC](https://forum.weaviate.io/t/any-plans-to-support-new-voyage-embeddings-or-jina-v2-reranker/3038/2 "2024-07-15T20:49:52Z")
hi @Baptiste_Cumin1 !! Welcome to our community! Is this voyage embedding the `voyage-multilingual-2` one? If that’s the case, good news: It is already supported. Here is how: ```python import weaviate import os from weaviate import classes as wvc client = weaviate.connect_to_local( headers={ "X-VoyageAI-Api-Key": os.getenv("VOYAGEAI_APIKEY"), } ) client.collections.delete("Voyage") collection = client.collections.create( "Voyage", vectorizer_config=wvc.config.Configure.Vectorizer.text2vec_voyageai( model="voyage-multilingual-2" ), properties=[ wvc.config.Property(name="some_text", data_type=wvc.config.DataType.TEXT) ] ) collection.data.insert( { "text": "This is an example" } ) ``` as for the Jina, indeed, we only have support for the text2vec: > **[Jina AI + Weaviate | Weaviate - Vector Database](https://weaviate.io/developers/weaviate/model-providers/jinaai)** > > Jina AI offers a wide range of models for natural language processing. Weaviate seamlessly integrates with Jina AI's APIs, allowing users to leverage Jina AI's models directly within the Weaviate database. I have raised this with our team and will keep you posted. Thanks for using Weaviate! ---
**Author:** ![DudaNogueira](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.weaviate.io/dudanogueira/32/7846_2.png) [@DudaNogueira](https://forum.weaviate.io/u/DudaNogueira)\ **Post date:** [July 15, 2024, 8:51pm UTC](https://forum.weaviate.io/t/any-plans-to-support-new-voyage-embeddings-or-jina-v2-reranker/3038/3 "2024-07-15T20:51:26Z")
Also, there is a PR pending review for the reranker from JinaAI: > > > \### What's being changed: > This PR adds a possibility of using JinaAi Reranker w…ith different RAG as: > Research was done to find out what reranker + Embeddings models work best using ; > Firstly many seed funded gen ai startups re confused which embeddings + reranker or rag to use so let simplify their problem if we can by giving some data > Embedding Models: OpenAI, CohereAI (v2.0/v3.0), Jina (small/base), BAAI/bge-large-en, Google PaLM, Voyage > Rerankers: CohereAI, bge-reranker-base, bge-reranker-large. > Hit Rate: Measures the fraction of queries where the correct answer is in the top-k retrieved documents. > Mean Reciprocal Rank (MRR): Evaluates accuracy based on the rank of the highest-placed relevant document. > These rerankers greatly refines search results improve hit rates and MRRs across embeddings > > Key Findings were-(1)-OpenaAI works best with cohere and bge-rerankers > (2)-JinaAi-Great with bge-reranker-large > But after JinaAi too launched there Jina Reranker (jina-reranker-v1-base-en) it achieves the highest among its peers with 0.8553 hit rate and 0.7091 mrr though the difference is not that much but still customers need the best for their companies > No Reranker jina-reranker bge--base bce-rer-base\_v1 cohere-reranker > Embedding model Hit Rate MRR Hit Rate MRR Hit Rate MRR Hit Rate MRR Hit Rate MRR > jina-v2-base-en 0.8053 0.5156 0.8737 0.7229 0.8368 0.6568 0.8737 0.7007 0.8842 0.7008 > bge-base-en-v1.5 0.7842 0.5183 0.8368 0.6895 0.8158 0.6586 0.8316 0.6843 0.8368 0.6739 > bce-embedding-base\_v1 0.8526 0.5988 0.8895 0.7346 0.8684 0.6927 0.9157 0.7379 0.9158 0.7296 > CohereV3-en 0.7211 0.4900 0.8211 0.6894 0.8000 0.6285 0.8263 0.6855 0.8316 0.6710 > Average 0.7908 0.5307 0.8553 0.7091 0.8303 0.6592 0.8618 0.7021 0.8671 0.6938 > !\[1\_tCBbIjV\_jLZP1AKLTX7rAw\](https://github.com/weaviate/weaviate/assets/112192260/b954e859-a89c-42d3-b44d-3b5057d6595d) > > > \### Review checklist > > \- \[x\] Documentation has been updated, if necessary. Link to changed documentation: > \- \[x\] Chaos pipeline run or not necessary. Link to pipeline: > \- \[x\] All new code is covered by tests where it is reasonable. > \- \[x\] Performance tests have been run or not necessary. ---
**Author:** ![DudaNogueira](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.weaviate.io/dudanogueira/32/7846_2.png) [@DudaNogueira](https://forum.weaviate.io/u/DudaNogueira)\ **Post date:** [July 24, 2024, 8:23pm UTC](https://forum.weaviate.io/t/any-plans-to-support-new-voyage-embeddings-or-jina-v2-reranker/3038/4 "2024-07-24T20:23:29Z")
Hi! We have a follow up! > > > \### What's being changed: > This PR adds a possibility of using JinaAi Reranker A…PI at https://api.jina.ai/v1/rerank. > > \### Review checklist > > \- \[x\] Documentation has been updated, if necessary. Link to changed documentation: > \- \[x\] Chaos pipeline run or not necessary. Link to pipeline: > \- \[x\] All new code is covered by tests where it is reasonable. > \- \[x\] Performance tests have been run or not necessary. Jina Reranker was added in [Release v1.26.1 - Hybrid search performance Fix, Tenants create API Fix, New JinaAI Reranker module · weaviate/weaviate · GitHub](https://github.com/weaviate/weaviate/releases/tag/v1.26.1) Thanks for using Weaviate!