Skip to content

DataEval/dingo

v2.4.0 Feature

This release adds 3 notable features for engineering teams evaluating rollout.

Published 5d MCP Data & Storage
✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

agent-as-a-judge common-crawl data-agent data-evaluation data-quality data-quality-assessment
+14 more
data-quality-report data-validation dataquality datascience deepseek hallucination hallucination-detection llm llm-as-a-judge openai opencompass qwen spark vlm

Summary

AI summary

Updates feat, fix, and llm across a mixed release.

Changes in this release

Feature Medium

Adds overall timeout to ArticleFactChecker and adds arXiv package.

Adds overall timeout to ArticleFactChecker and adds arXiv package.

Source: llm_adapter@2026-05-29

Confidence: high

 —
Feature Medium

Adds raw case handling with dingo_id.

Adds raw case handling with dingo_id.

Source: llm_adapter@2026-05-29

Confidence: high

 —
Feature Medium

Adds mineru format support.

Adds mineru format support.

Source: llm_adapter@2026-05-29

Confidence: high

 —
Feature Medium

Adds User-Agent header to image download.

Adds User-Agent header to image download.

Source: llm_adapter@2026-05-29

Confidence: high

 —
Feature Medium

Adds configurable LLMCustomRule evaluator.

Adds configurable LLMCustomRule evaluator.

Source: llm_adapter@2026-05-29

Confidence: low

 —
Feature Medium

Adds LLMHtmlExtractCompareV3 metric.

Adds LLMHtmlExtractCompareV3 metric.

Source: llm_adapter@2026-05-29

Confidence: low

 —
Feature Low

Add clawhub skill badge.

Add clawhub skill badge.

Source: granite4.1:30b@2026-05-29-audit

Confidence: low

 —
Feature Low

Add table and equation quality prompts.

Add table and equation quality prompts.

Source: granite4.1:30b@2026-05-29-audit

Confidence: low

 —
Feature Low

Make BaseTextQuality judge pass/fail based on score.

Make BaseTextQuality judge pass/fail based on score.

Source: granite4.1:30b@2026-05-29-audit

Confidence: low

 —
Feature Low

Add RuleDictConsistency check (order‑insensitive).

Add RuleDictConsistency check (order‑insensitive).

Source: granite4.1:30b@2026-05-29-audit

Confidence: low

 —
Bugfix Medium

Fixes LLMHtmlExtractCompareV2 metric.

Fixes LLMHtmlExtractCompareV2 metric.

Source: llm_adapter@2026-05-29

Confidence: high

 —
Bugfix Medium

Updates LLMHtmlExtractCompareV3 response type.

Updates LLMHtmlExtractCompareV3 response type.

Source: llm_adapter@2026-05-29

Confidence: high

 —
Bugfix Medium

Injects LLM config into class in local executor.

Injects LLM config into class in local executor.

Source: llm_adapter@2026-05-29

Confidence: high

 —
Bugfix Medium

Optimizes RequiredField sort.

Optimizes RequiredField sort.

Source: llm_adapter@2026-05-29

Confidence: low

 —
Bugfix Low

Optimize RequiredField sort algorithm.

Optimize RequiredField sort algorithm.

Source: granite4.1:30b@2026-05-29-audit

Confidence: low

 —
Refactor Low

Remove parameters attribute from EvaluatorLLMArgs to allow extension.

Remove parameters attribute from EvaluatorLLMArgs to allow extension.

Source: granite4.1:30b@2026-05-29-audit

Confidence: low

 —
Full changelog

What's Changed

  • feat: add clawhub skill badge by @e06084 in https://github.com/MigoXLab/dingo/pull/374
  • feat(llm): add overall timeout to ArticleFactChecker & add arXiv package by @seancoding-day in https://github.com/MigoXLab/dingo/pull/375
  • fix: LLMHtmlExtractCompareV2 metric by @e06084 in https://github.com/MigoXLab/dingo/pull/376
  • feat: table、equation质检prompt by @shijinpjlab in https://github.com/MigoXLab/dingo/pull/377
  • docs: update wechat by @e06084 in https://github.com/MigoXLab/dingo/pull/378
  • feat: BaseTextQuality根据score来判断 by @shijinpjlab in https://github.com/MigoXLab/dingo/pull/379
  • metric: update LLMTextQualityV5 by @e06084 in https://github.com/MigoXLab/dingo/pull/380
  • feat: EvaluatorLLMArgs删除parameters属性，允许扩展 by @shijinpjlab in https://github.com/MigoXLab/dingo/pull/381
  • docs: update wechat (#378) by @shijinpjlab in https://github.com/MigoXLab/dingo/pull/382
  • feat: add LLMHtmlExtractCompareV3 metric by @e06084 in https://github.com/MigoXLab/dingo/pull/384
  • fix: optimize RequiredField sort by @e06084 in https://github.com/MigoXLab/dingo/pull/385
  • docs: readme add saas screenshot by @e06084 in https://github.com/MigoXLab/dingo/pull/386
  • feat: RuleDictConsistency by @shijinpjlab in https://github.com/MigoXLab/dingo/pull/387
  • docs: add mineru xunlianying doc by @e06084 in https://github.com/MigoXLab/dingo/pull/388
  • fix: update LLMHtmlExtractCompareV3 response type by @e06084 in https://github.com/MigoXLab/dingo/pull/389
  • feat: RuleDictConsistency忽略顺序 by @shijinpjlab in https://github.com/MigoXLab/dingo/pull/391
  • update error judge by @dt-yy in https://github.com/MigoXLab/dingo/pull/393
  • feat: raw情况添加dingo_id by @shijinpjlab in https://github.com/MigoXLab/dingo/pull/394
  • sync main to dev by @e06084 in https://github.com/MigoXLab/dingo/pull/398
  • feat(llm): add configurable LLMCustomRule evaluator by @daniel5u in https://github.com/MigoXLab/dingo/pull/397
  • feat: llm custom metric by @daniel5u in https://github.com/MigoXLab/dingo/pull/401
  • fix: inject llm config into class in local executor by @daniel5u in https://github.com/MigoXLab/dingo/pull/403
  • Feature/lld add chunk test for rag by @pekopoke in https://github.com/MigoXLab/dingo/pull/406
  • Dev scibase by @shijinpjlab in https://github.com/MigoXLab/dingo/pull/405
  • feat: unify image loader by @e06084 in https://github.com/MigoXLab/dingo/pull/407
  • feat: image download add User-Agent by @e06084 in https://github.com/MigoXLab/dingo/pull/408
  • docs: update wechat by @e06084 in https://github.com/MigoXLab/dingo/pull/410
  • feat: add mineru format support by @e06084 in https://github.com/MigoXLab/dingo/pull/412
  • Release v2.4.0 by @e06084 in https://github.com/MigoXLab/dingo/pull/413
  • Release v2.4.0 by @e06084 in https://github.com/MigoXLab/dingo/pull/414

Full Changelog: https://github.com/MigoXLab/dingo/compare/v2.2.1...v2.4.0

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track DataEval/dingo

Get notified when new releases ship.

Sign up free

About DataEval/dingo

MCP server for the Dingo: a comprehensive data quality evaluation tool. Server Enables interaction with Dingo's rule-based and LLM-based evaluation capabilities and rules&prompts listing.

All releases →

Related context

Earlier breaking changes

  • v2.3.0 Releases version 2.3.0.

Beta — feedback welcome: [email protected]