---
title: "Discussion on \"Late Chunking + Jina v4: Context-Aware Embeddings, Self-Hosted on GCP\" | Hashnode"
description: "Discussion on \"Late Chunking + Jina v4: Context-Aware Embeddings, Self-Hosted on GCP\". How We Fixed RAG Chunking — and Moved Off HuggingFace Spaces"
url: "https://hashnode.com/posts/late-chunking-jina-v4-context-aware-embeddings-self-hosted-on-gcp/6a8d507e6234228829f0b857"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [Late Chunking + Jina v4: Context-Aware Embeddings, Self-Hosted on GCP](https://zeonai.hashnode.dev/late-chunking-jina-v4-context-aware-embeddings-self-hosted-on-gcp?utm_source=hashnode&utm_medium=feed)

How We Fixed RAG Chunking — and Moved Off HuggingFace Spaces

Aug 25 · 7 min read · zeonai.hashnode.dev  
Read the article

-   [#rag](https://hashnode.com/tag/rag)
-   [#llm](https://hashnode.com/tag/llm)
-   [#embeddings](https://hashnode.com/tag/embeddings)
-   [#gcp](https://hashnode.com/tag/gcp)
-   [#kubernetes](https://hashnode.com/tag/kubernetes)

## Written by

[Pradipta Nanda](https://hashnode.com/@sp1dy)

@sp1dy · Bangalore, India

  
[View profile](https://hashnode.com/@sp1dy)

Software development Engineer 2

## Responses 3

Markdown supported · ⌘↵ to post

[Duko tools](https://hashnode.com/@talha24ramzan)

[Aug 27](https://hashnode.com/posts/late-chunking-jina-v4-context-aware-embeddings-self-hosted-on-gcp/6a8d507e6234228829f0b857/comment/6a8fd34114b74994b654cafd)

"Embed first, chunk later" is the right framing, most teams treat chunking as a one-time tuning step when it's actually deciding how much context survives retrieval.

The honest cost table stands out most though, most migration writeups only show numbers that justify the move. Admitting HF Spaces still wins on raw cost for bursty workloads makes the self-hosting case land harder, since it's clearly not being oversold as cheaper.

[Sanoary Istiaque](https://hashnode.com/@istyy)

[Aug 26](https://hashnode.com/posts/late-chunking-jina-v4-context-aware-embeddings-self-hosted-on-gcp/6a8d507e6234228829f0b857/comment/6a8ebecbb16cbfc03543ddaa)

Really interesting read! I liked how the post connects late chunking with context-aware embeddings and then takes it into a practical self-hosted GCP setup. The explanation makes a fairly technical topic feel much more approachable. The trade-offs around preserving context while keeping retrieval efficient were especially interesting. Definitely a useful read for anyone working with RAG and modern embedding pipelines.

[Kartik N V J K](https://hashnode.com/@kartik-nvj)

[Aug 25](https://hashnode.com/posts/late-chunking-jina-v4-context-aware-embeddings-self-hosted-on-gcp/6a8d507e6234228829f0b857/comment/6a8dfc085dca1240fc336c27)

The late-chunking flip is the part I keep coming back to, embedding first and cutting after preserves neighbor context that vanilla chunk-then-embed silently drops. I also like that you kept the latency numbers honest (0.14 to 0.8s per chunk on self-hosted GCP vs 1 to 3s on Spaces), the "always-on cost" caveat is what most write-ups skip. Curious whether you saw retrieval quality shift on tables and code blocks specifically, those tend to break mean-pooled token embeddings for us.