Blog

Engineering

How PlanetScale Powers the Mixedbread Control Plane

How PlanetScale Powers the Mixedbread Control Plane

Why Mixedbread moved to PlanetScale for database reliability, low latency, observability, and hands-on support.

EngineeringProduct
4 min readAugust 14, 2026
Building a Robust Ingestion System for Any File of Any Size

Building a Robust Ingestion System for Any File of Any Size

A 300GB video does not fit on one machine, and a 4MB PDF can render into gigabytes of images. This is how our slicer and parser architecture processes arbitrary files with bounded compute and memory.

Engineering
13 min readJuly 14, 2026
Asymmetric Quantization: Near-Lossless Late Interaction Retrieval with 97% Storage Reduction

Asymmetric Quantization: Near-Lossless Late Interaction Retrieval with 97% Storage Reduction

Late interaction makes retrieval more precise but turns every document into many vectors. Asymmetric quantization keeps query vectors precise and stores document vectors as binary signs, cutting corpus storage 32x while holding 89.65 vs 90.26 NDCG@10.

ResearchEngineering
8 min readJune 29, 2026
Closing the Oracle Gap for Your Agents

Closing the Oracle Gap for Your Agents

Mixedbread Search v3 narrows the oracle gap for agentic retrieval, topping BrowseComp-Plus and setting leading results on MADQA and OfficeQA-Pro.

EngineeringResearchProduct
7 min readMarch 24, 2026
Beyond the Limit: Introduce Mixedbread Wholembed v3

Beyond the Limit: Introduce Mixedbread Wholembed v3

Announcing Mixedbread Wholembed v3, our new unified omnimodal multilingual retrieval model, setting a new state of the art for search across languages, modalities, and real-world retrieval tasks.

EngineeringResearchProduct
8 min readMarch 12, 2026
Inside Mixedbread: How We Built Multimodal Late-Interaction at Billion Scale

Inside Mixedbread: How We Built Multimodal Late-Interaction at Billion Scale

Technical deep-dive into Mixedbread Search - the first production-ready late-interaction search with native multimodality. Learn how we achieve ~80ms end-to-end latency on billion-scale document collections.

EngineeringResearchProduct
12 min readJanuary 21, 2026
maxsim-cpu: Maximising Maxsim Efficiency

maxsim-cpu: Maximising Maxsim Efficiency

Introducing maxsim-cpu, a much faster way to compute the late interaction's MaxSim operator on modern CPU hardware, optimised for both x86 and Mac ARM.

ResearchEngineering
11 min readJuly 15, 2025
Baking in Performance - Dynamic Batching with Batched

Baking in Performance - Dynamic Batching with Batched

Learn how dynamic batching boosts GPU efficiency in machine learning. We introduce Batched, our open-source library that makes it easy to implement dynamic batching and improve performance in your ML projects.

Engineering
8 min readSeptember 16, 2024
Getting Better with Baguetter - New Retrieval Testing Framework

Getting Better with Baguetter - New Retrieval Testing Framework

Baguetter is our new open-source retrieval experimentation framework that combines sparse, dense, and hybrid retrieval into a single, easy-to-use interface. It enables fast benchmarking, implementation, and testing of new search methods.

Engineering
7 min readAugust 23, 2024
BM𝒳: A Freshly Baked Take on BM25

BM𝒳: A Freshly Baked Take on BM25

Introducing BMX, an iteration on the industry standard BM25 search algorithm. Through the incorporation of entropy-weighted query-document similarity and weighted query augmentation, the algorithm can increase search performance on the most relevant information retrieval benchmarks.

ResearchEngineering
15 min readAugust 12, 2024