quickwit-oss/quickwit

Health v2 · maintenance only

Adoption not part of the score

11549 stars · 582 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Quickwit is an open-source, cloud-native search engine written in Rust and powered by Tantivy, designed to run sub-second search and analytics queries directly on cloud object storage like S3. It is optimized for observability workloads such as log management and distributed tracing, with native OpenTelemetry and Jaeger support and an Elasticsearch-compatible API.

Use cases

• search and analyze application logs stored on S3
• backend trace storage for Jaeger
• ingest OpenTelemetry logs and traces
• replace Elasticsearch for low-QPS, high-volume log search
• full-text search over petabytes of immutable data
• query logs with an Elasticsearch-compatible API
• self-hosted log management on Kubernetes

When to choose

• you need cost-efficient full-text search over large volumes of append-only data on object storage
• you want a Jaeger or OpenTelemetry-native traces backend
• you need an Elasticsearch-compatible API without running an Elasticsearch cluster
• you want decoupled compute and storage with sub-second search latency

When to avoid

• you need high-QPS search over frequently updated data, which traditional engines handle better
• you need metrics storage today, as support is still on the roadmap
• you require Windows support, which is not available
• you need frequent updates or deletions of indexed documents

Facets

service · maturity active

search-engine http-server api-framework cli logging tracing monitoring big-data monitoring cloud-computing self-hosted analytics self-hosted cloud rust cli observability log-management distributed-tracing opentelemetry jaeger elasticsearch-compatible tantivy object-storage cloud-native full-text-search search devops linux macos docker kubernetes

Member repositories

For agents

markdown · JSON · MCP: product_card(name="quickwit-oss/quickwit")