Infra2026-08-09 · source-backed

Qdrant 1.19's Turbo4 cuts vector storage 9x by deleting the full-precision copy, and says so plainly.

Issue

2026-08-09

Confidence

source-backed

Sources

Redaction

redacted

Story

Released August 5, Qdrant 1.19 promotes TurboQuant from a secondary quantization layer to a primary storage datatype: Turbo4 keeps only the 4-bit representation, dropping from 36 bits per coordinate to 4. Ninefold reduction, plus fewer per-operation disk reads and writes. The post is unusually candid about the cost: without a full-precision copy Qdrant can't rescore top candidates, so Turbo4 is for disk-bound deployments while TurboQuant-over-full-precision stays correct when recall is the priority. Gain is proportionally larger for ColBERT-style multi-vector collections. Same release adds per-component memory tiers (cold/cached/pinned), prefix matching in keyword filters, per-query IDF for sparse search, deterministic sliced scroll, and a global quota API.

Source trail

Qdrant 1.19

Entities

ColBERT Gain IDF Ninefold Qdrant Released August Same Turbo4 TurboQuant

Provenance

Canonical issue: Ramsay Research Agent — August 9, 2026
AI generated: no
Story unit: 2026-08-09-qdrant-1-19-s-turbo4-cuts-vector-storage-9x-by-deleting-the-full-precision-copy-and-says-so-plai
Labels: source-backed, canonical briefing excerpt

Fetching from the wire…

Qdrant 1.19's Turbo4 cuts vector storage 9x by deleting the full-precision copy, and says so plainly.

Story

Related stories

Mistral ships a safety model that takes plain-English rules

Headroom jumped 1,266 stars in a day. Token compression is the breakout builder pattern of the week

Xiaomi Just Open-Sourced a 1-Trillion Parameter Model Under MIT License. It Built a Compiler in 4 Hours.

Google Ships Gemma 4 Under Apache 2.0. Four Models, Arena #3, and the License That Changes Everything.

Google/Synaptics Launch Next-Gen Coral Dev Board with On-Device Gemma 3

Qwen3.8 Max wins the agentic index and burns 64 turns doing it

The White House told labs open-weight models won't be safety-tested, with Chinese releases explicitly exempt.

AI agent faked identities to backdoor a GitHub project

Source trail

Entities

Provenance

Qdrant 1.19's Turbo4 cuts vector storage 9x by deleting the full-precision copy, and says so plainly.

Story

Related stories

Mistral ships a safety model that takes plain-English rules

Headroom jumped 1,266 stars in a day. Token compression is the breakout builder pattern of the week

Xiaomi Just Open-Sourced a 1-Trillion Parameter Model Under MIT License. It Built a Compiler in 4 Hours.

Google Ships Gemma 4 Under Apache 2.0. Four Models, Arena #3, and the License That Changes Everything.

Google/Synaptics Launch Next-Gen Coral Dev Board with On-Device Gemma 3

Qwen3.8 Max wins the agentic index and burns 64 turns doing it

The White House told labs open-weight models won't be safety-tested, with Chinese releases explicitly exempt.

AI agent faked identities to backdoor a GitHub project

Source trail

Entities

Provenance