# Meilisearch 1.55 fixes ranking-threshold slowdowns and blocks a vector-index corruption path

The October 12 release stops rankingScoreThreshold queries from over-scanning, rejects a previously accepted binaryQuantization:null setting that could corrupt embeddings, and adds an experimental Enterprise-only search-rule preview.

A search-engine maintenance release carries intentional relevance and API compatibility changes that operators should test.

- Status: Active
- Published: 2026-10-13T06:18:24+13:00
- Updated: 2026-10-13T06:18:24+13:00
- Categories: Web Development, Cloud & Infrastructure, Developer Tools, Databases & Storage
- Tags: databases, Meilisearch, open source, search relevance
- Canonical HTML: https://beyondthe.news/dossiers/meilisearch-155-ranking-threshold-quantization-api-fixes

## What changed

Meilisearch 1.55.0, released October 12, 2026, changes rankingScoreThreshold processing to stop after the first below-threshold document, avoiding potentially evaluating every document. Maintainers warn relevance can change. The API now rejects binaryQuantization:null when updating embedder settings because turning off binary quantization after discarding original vectors could corrupt embeddings. It also adds Enterprise-only experimental inlineRule previews and fixes a Unicode highlighting panic.

## Why it matters

Teams using Meilisearch for ecommerce, documentation and vector-backed search should test result quality and latency during upgrades. Settings clients that use null to reset binaryQuantization may now receive a deliberate error. Enterprise rule-preview functionality is useful for merchandising experiments but is not stable or generally available.

## Ranking thresholds now stop earlier

rankingScoreThreshold formerly could evaluate far too many documents, sometimes the whole index. Version 1.55 stops at the first document below the score threshold. Maintainers explicitly warn that result relevance may change in some circumstances; compare real queries before rollout.

## The quantization reset now fails safely

Binary quantization discards exact original embedding values. The old API accidentally allowed binaryQuantization:null, effectively disabling it and risking corrupted embeddings. Version 1.55 rejects this update. Check settings-reset code and handle the new validation error.

## Search-rule previews are experimental and commercial

inlineRule lets Enterprise Edition users with experimental dynamicSearchRules test a temporary rule in search or multi-search without persisting it. The Meili-Include-Metadata header can explain condition outcomes. This is not a broadly stable open-source feature.

## Unicode highlighting crash is fixed

Certain escaped Unicode code points could trigger a panic during highlight formatting with some segmenters. The patch fixes that path; the affected scope depends on index content and configuration.

## Upgrade guidance

Benchmark representative threshold queries for both latency and result order, remove unsupported binaryQuantization resets, and check edition/feature flags before using inlineRule. No independent production performance numbers were published.

## Key details

- Meilisearch v1.55.0 published October 12, 2026.
- rankingScoreThreshold now stops after first below-threshold document.
- Maintainers warn that the fix can alter search relevance in some cases.
- binaryQuantization:null now fails rather than risking corrupted embeddings.
- inlineRule requires Enterprise Edition and experimental dynamicSearchRules.
- A Unicode highlighting panic is fixed.

## Builder takeaways

- Regression-test search relevance and latency when upgrading.
- Audit embedder settings for binaryQuantization:null.
- Do not assume inlineRule is stable or available in the open-source edition.
- Check Unicode highlighting behaviour on representative data.

## What to watch

- Rule preview stabilisation and Cloud UI availability.
- Relevance regressions from the threshold change.
- Migration guidance for previously corrupted quantised indexes.

## Uncertainties

- No independent performance benchmark or affected-installation count was verified.
- Relevance effects depend on index configuration and query patterns.
- Experimental rule-preview API may change.

## Sources

- [Meilisearch v1.55.0 release](https://github.com/meilisearch/meilisearch/releases/tag/v1.55.0) — Meilisearch maintainers · primary · 2026-10-12T00:00:00+13:00. Authoritative release changes and compatibility warnings.
- [Binary quantization API guard PR #6656](https://github.com/meilisearch/meilisearch/pull/6656) — Meilisearch contributors · primary · 2026-10-12T00:00:00+13:00. Underlying fix rationale.
- [Meilisearch v1.55.0 release mirror](https://newreleases.io/project/github/meilisearch/meilisearch/release/v1.55.0) — NewReleases · independent · 2026-10-12T00:00:00+13:00. Mirrors version metadata; not independent testing.