Top all-time · #5546
Voyage AI
Company blog · blog.voyageai.comaidata
Voyage AI is the company blog at blog.voyageai.com, writing about ai and data. Since 2023, 14 different Hacker News readers have linked it 116 times, most often “voyage-context-3: Contextual Retrieval Without the LLM”. It is still writing: the last post was on 2026-10-28, and you can follow it by feed.
Too few sampled comments (4) to say how often HN recommends it.?
Why HN links it
HN readers link Voyage AI's posts announcing embedding and multimodal retrieval models, plus its MongoDB acquisition, as the source for new model releases and their details.
Summarised by Claude from a sample of 4 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Most-linked posts?
- voyage-context-3: Contextual Retrieval Without the LLM5 people · 6 mentions · first linked 2025-07-23
- The Voyage 4 model family: shared embedding space with MoE architecture3 people · 3 mentions · first linked 2026-01-15
- Voyage-3-large: the new general-purpose embedding model2 people · 18 mentions · first linked 2025-01-08
- 32k context length text embedding models2 people · 13 mentions · first linked 2024-09-23
- Voyage-multimodal-3.5: retrieval frontier with video support2 people · 5 mentions · first linked 2026-01-15
- All-in-one embedding model for interleaved text, images, and screenshots2 people · 4 mentions · first linked 2024-11-12
- Voyage Embeddings2 people · 2 mentions · first linked 2023-10-31
- Voyage AI Joins MongoDB2 people · 2 mentions · first linked 2025-02-24
- Voyage AI Raises $28M2 people · 2 mentions · first linked 2024-10-03
What Hacker News says?
We provide retrieval metrics for a variety of datasets and languages: https://blog.voyageai.com/2024/09/18/voyage-3/ . I also personally encourage folks to either test on their own data or to find an open source dataset that closely resembles the documents they are trying to search (we provide a ton of free tokens for the evaluating our models).
Comment linking /2024/09/18/voyage-3/
on “Embeddings are underrated”Submitted All-in-one embedding model for interleaved text, images, and screenshots as a story · 263 points
HN discussionIt is unclear this model should be on that leaderboard because we don't know whether it has been trained on mteb test data. It is worth noting that their own published material [0] does not entail any score from any dataset from the mteb benchmark. This may sound nit picky, but considering transformers' parroting capabilities, having seen test data during training should be expected to completely invalidate those sc…
Comment linking /2024/09/18/voyage-3/
on “Embeddings are underrated”Submitted Voyage-code-3 as a story · 111 points
HN discussionGlad you liked the results! We do have multilingual models (and rerankers) -- voyage-3, in particular, is multilingual: https://blog.voyageai.com/2024/09/18/voyage-3/ voyage-multimodal-3 is multilingual as well, supporting the same set of languages as voyage-3.
Comment linking /2024/09/18/voyage-3/
on “All-in-one embedding model for interleaved text, images, and screenshots”Submitted 32k context length text embedding models as a story · 101 points
HN discussionVoyage AI post: https://blog.voyageai.com/2025/02/24/joining-mongodb/
Comment linking /2025/02/24/joining-mongodb/
on “MongoDB acquires Voyage AI”Submitted Rerank-2.5 and rerank-2.5-lite: instruction-following rerankers as a story · 93 points
HN discussionFor float and int8, 1024 does indeed outperform 2048. However, for binary, 2048 outperforms 1024. For our general-purpose embedding model (voyage-3-large), embedding vectors with 2048 dimensions outperform 1024 across the board: https://blog.voyageai.com/2025/01/07/voyage-3-large/
Comment linking /2025/01/07/voyage-3-large/
on “Voyage-code-3”Submitted How do we evaluate vector-based code retrieval? as a story · 55 points
HN discussionCool to see this but also check out voyage-4 embeddings including the open weights nano https://blog.voyageai.com/2026/01/15/voyage-4/ The fact that the different sizes on the 4 series share an embeddings space is awesome
Comment linking /2026/01/15/voyage-4/
on “voyage-multimodal-3.5: a new multimodal retrieval frontier with video support”Submitted voyage-3.5 and voyage-3.5-lite: improved quality for a new retrieval frontier as a story · 32 points
HN discussionNot yet, we just saw their announcement today here (for context): https://blog.voyageai.com/2023/10/29/voyage-embeddings/ . We'll definitely work on adding this model next. Seems promising! Thanks for sharing.
Comment linking /2023/10/29/voyage-embeddings/
on “Show HN: Playground for comparing embedding models on Wikipedia+book retrieval”Submitted Evaluating Code Embedding Models as a story · 25 points
HN discussionSubmitted voyage-context-3: Contextual Retrieval Without the LLM as a story · 7 points
HN discussionSubmitted Voyage-context-3: focused chunk-level details with global document context as a story · 4 points
HN discussionSubmitted Voyage AI Joins MongoDB as a story · 4 points
HN discussionSubmitted Voyage-3-large: the new general-purpose embedding model as a story · 4 points
HN discussionSubmitted Create your own Jev using Voyage models as a story · 3 points
HN discussionSubmitted Voyage-code-4: code retrieval built for coding agents as a story · 3 points
HN discussionSubmitted Voyage-multimodal-3.5: retrieval frontier with video support as a story · 3 points
HN discussionSubmitted voyage-context-3: Contextual Retrieval Without the LLM as a story · 3 points
HN discussionSubmitted voyage-3-Large as a story · 3 points
HN discussionSubmitted voyage-3-large as a story · 3 points
HN discussionSubmitted 32K-token context length embedding models as a story · 3 points
HN discussionSubmitted The Voyage 4 model family: shared embedding space with MoE architecture as a story · 2 points
HN discussionSubmitted The Case Against LLMs as Rerankers as a story · 2 points
HN discussionSubmitted The Case Against LLMs as Rerankers as a story · 2 points
HN discussionSubmitted VoyageAI's context-aware embeddings improve document-level similarity search as a story · 2 points
HN discussionSubmitted Voyage-context-3: focused chunk-level details with document context as a story · 2 points
HN discussion
