RAG и системы знаний

FastGPT

labring/FastGPT

A visual knowledge-workflow platform with document ingestion, hybrid retrieval, reranking, and application publishing.

★ 29,6KЗвёзды
⑂ 7,3KФорки
187Открытые проблемы
TypeScriptЯзык
Лицензия
Q90Редакционная оценка

Обзор

FastGPT combines knowledge preparation, model calls, and delivery in configurable workflows. Its search tests, editable chunks, citations, and call traces help separate retrieval failures from generation failures. It uses a custom license with material SaaS restrictions.

Основные функции

  • Visual workflows
  • Document ingestion
  • Hybrid search
  • Retrieval testing
  • Embedding and logs

Требования, установка и быстрый старт

Use the version-matched official Docker Compose files and configure storage, model endpoints, and administration before starting services.

Использование

Test exact facts, ambiguous terms, and missing answers against one manual. Validate retrieval before tuning generation and inspect citations and call logs.

How it works
Documents become searchable chunks. Workflow nodes retrieve, rerank, call models or tools, and deliver answers through an embedded UI or API.

Audience and requirements
Product and engineering teams building knowledge apps. Containers, storage/search dependencies, chat and embedding models.

Practical use cases
Product documentation assistants; internal workflows; embedded knowledge chat.

Limitations and selection
Complex flows need engineering. Index freshness and access boundaries require maintenance. Review the custom license before a SaaS offering.

Related projects and selection
onyx-dot-app/onyx:Comparison: assess Onyx for connector synchronization and FastGPT for explicit workflow composition.

google/langextract:Possible complement: extract structured fields before feeding a custom knowledge workflow.

Source review
Editorial analysis of upstream sources, without runtime or benchmark testing. Proposed workflows are editorial suggestions.

Совместимость моделей и варианты использования

Configure and validate chat, embedding, and reranking models separately.

Лицензия и примечания о рисках

Uses the FastGPT Open Source License. The README allows commercial backend use but disallows SaaS provision, and requires copyright notices without commercial authorization. Read the complete LICENSE.

Editorial source review 2026-09-09T05:00:00.950Z. README and live repository page verified; current stars/forks from GitHub HTML. Last-push metadata retained from 2026-09-05 discovery snapshot. No runtime benchmark. Integration proposals are editorial analysis.

Релиз и сопровождение

Reviewed 2026-09-09. Counters come from repository pages; features are based on upstream documentation. See Releases in the source links. Editorial analysis of upstream sources, without runtime or benchmark testing. Proposed workflows are editorial suggestions.

milvus

milvus-io/milvus

★ 45,7KGo

LanceDB

lancedb/lancedb

★ 11,1KRust

EvalScope

modelscope/evalscope

★ 3,2KPython

llm-app

pathwaycom/llm-app

★ 58,9KJupyter Notebook

graphrag

microsoft/graphrag

★ 35,9KPython

onyx

onyx-dot-app/onyx

★ 32KPython