NewYour coding agent can read the release notes before it upgrades.Set up the MCP server →
Go modules · #5 by repository stars
Last release 7 months ago
10 Feb 2026
Ships fairly regularly
a new release about every 2 weeks
Most releases are documented
notes for 32 of 39 stable releases
Nothing withdrawn
no release was ever pulled
2 years old
85 releases · first in 2024
Nothing published for this version
Introduces memory management APIs (HTTP and Python).
Released on February 10, 2026.
One column per month.
Nothing published for this version
Memory: Enhances the stability of memory extraction when all memory types are selected.
Released on December 31, 2025.
Implements a Memory interface for managing memory.
Released on December 27, 2025.
Supports exporting Agent outputs in Word or Markdown formats.
Released on November 19, 2025.
-full RAGFlow edition.Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
:::danger IMPORTANT From this release onwards, we ship only the slim edition (without embedding models) Docker image and no longer append the -slim su
Released on November 12, 2025.
:::danger IMPORTANT
From this release onwards, we ship only the slim edition (without embedding models) Docker image and no longer append the -slim suffix to the image tag.
:::
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Experimental: Adds support for PDF document parsing using MinerU. See here.
Nothing published for this version
Orchestratable ingestion pipeline: Supports customized data ingestion and cleansing workflows, enabling users to flexibly design their data flows or d
Released on October 15, 2025.
Released on September 10, 2025.
Released on September 10, 2025.
"metadata_condition" to the Retrieve chunks method, enabling metadata-based chunk filtering during retrieval. #9877metadata_condition to the Retrieve chunks method, enabling metadata-based chunk filtering during retrieval. #9877Nothing published for this version
Agent component: Completes Chinese localization for the Agent component.
Released on August 27, 2025.
ENABLE_TIMEOUT_ASSERTION environment variable to enable or disable timeout assertions for file parsing tasks.ZHIPU GLM-4.5
Ecommerce Customer Service Workflow: A template designed to handle enquiries about product features and multi-product comparisons using the internal dataset, as well as to manage installation appointment bookings.
Nothing published for this version
Nothing published for this version
Nothing published for this version
Revamps the user interface for the Datasets, Chat, and Search pages.
Released on August 20, 2025.
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
The Retrieval component now supports the dynamic specification of dataset names using variables.
Released on August 8, 2025.
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
From v0.20.0 onwards, Agents are no longer compatible with earlier versions, and all existing Agents from previous versions must be rebuilt following
Released on August 4, 2025.
From v0.20.0 onwards, Agents are no longer compatible with earlier versions, and all existing Agents from previous versions must be rebuilt following the upgrade.
A memory leak issue during high-concurrency requests.
Released on June 23, 2025.
Cross-language search is supported in the Knowledge and Chat modules, enhancing search accuracy and user experience in multilingual environments, such
Released on May 26, 2025.
The following features have been contributed by our community:
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
From this release onwards, built-in rerank models have been removed because they have minimal impact on retrieval rates but significantly increase ret
Released on April 23, 2025.
From this release onwards, built-in rerank models have been removed because they have minimal impact on retrieval rates but significantly increase retrieval time.
Removes the Max_tokens setting from Chat configuration.
Released on March 13, 2025.
From this release onwards, if you still see RAGFlow's responses being cut short or truncated, check the Max_tokens setting of your model provider.
Improves English tokenization quality.
Released on March 11, 2025.
Nothing published for this version
AI chat: Implements Deep research for agentic reasoning. To activate this, enable the Reasoning toggle under the Prompt engine tab of your chat assist
Released on March 3, 2025.
/ can be used to insert available keys (variables) in the system prompt field of the Generate or Template component.The following diagram illustrates the workflow of RAGFlow's Deep research:
The following is a screenshot of a conversation that integrates Deep research:
Adds a body parameter "meta_fields" to the Update document method.
Adds a key option "meta_fields" to the Update document method.
Nothing published for this version
Nothing published for this version
Supports DeepSeek R1 and DeepSeek V3.
Released on February 6, 2025.
:::danger IMPORTANT The Tag dataset feature is unavailable on the Infinity document engine. :::
Upgrades RAGFlow's document engine Infinity to v0.5.2.
Released on December 25, 2024.
This release fixes the following issues:
SCORE not found and position_int errors returned by Infinity.Adds an optional parameter "user_id" to the following APIs:
Nothing published for this version
Introduces additional Agent-specific APIs.
Released on December 18, 2024.
Nothing published for this version
Your coding agent can read these notes before it upgrades. Set up the MCP server →