SuperCLUE-RAG
Chinese models answer questions twice, first without documents and then with supplied documents. A judge model compares the answers to rate how well the system uses retrieved text.
The repository has about 133 stars. It also includes questions with no documents.
- Level
- Everyday Most adults, without training
- Scored
- Graded by a model A language model scores each response, usually against a rubric.
- Answers
- Open-ended Free-form text, documents, designs or media judged as a whole.
Mi-BRAG tops the leaderboard with 78.1
Best score per entry from the benchmark's own leaderboard, read on 2 Nov 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is GPT-4o, at 66.7.
- 1 Mi-BRAG 78.1
- 2 360智脑 75.4
- 3 GLM-4-0520 71.5
- 4 MiniMax-abab6.5s-chat 68.3
- 5 通义千问2.5 68.0
- 6 文心一言4-turbo 67.0
- 7 GPT-4o May 2024 66.7
- 8 文心一言3.5 66.4
- 9 讯飞星火4.0 65.5
- 10 Moonshot-v1-128k 57.4
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Mi-BRAG | 78.1 | Apr 2025 | Official leaderboard | |
| 2 | 360智脑 | 75.4 | Nov 2024 | Official leaderboard | |
| 3 | GLM-4-0520 | 71.5 | Nov 2024 | Official leaderboard | |
| 4 | MiniMax-abab6.5s-chat | 68.3 | Nov 2024 | Official leaderboard | |
| 5 | 通义千问2.5 | 68.0 | Nov 2024 | Official leaderboard | |
| 6 | 文心一言4-turbo | 67.0 | Nov 2024 | Official leaderboard | |
| 7 | GPT-4o May 2024 | 66.7 | May 2024 | Official leaderboard | |
| 8 | 文心一言3.5 | 66.4 | Nov 2024 | Official leaderboard | |
| 9 | 讯飞星火4.0 | 65.5 | Nov 2024 | Official leaderboard | |
| 10 | Moonshot-v1-128k | 57.4 | Nov 2024 | Official leaderboard |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
No lab model card reports it yet
We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark.
More in Factuality
- SimpleVQA In use Models answer 2,025 short factual questions about images across 9 tasks and 9 topics. An LLM judge grades answers against fixed references.
Qwen3.7-Plus 81.7% - HallusionBench In use Models answer 1,129 yes or no questions about 346 images, built in pairs to expose language hallucination and visual illusion. Scored by accuracy.
Keye-VL-2.0-30B-A3B 73.5% - SimpleQA Verified In use 1,000 human-verified, de-duplicated short-answer factual questions measuring parametric knowledge.
Gemini 3.1 Pro 77.5% - GaRAGe In use Models write long-form RAG answers to 2,366 questions with human-annotated grounding passages. Scoring checks whether answers use the relevant passages.
Nemotron 3.5 Bolt E4B 40.7% - Kimi Hallu Bench In use Models answer hard prompts drawn from practical work tasks. Fact-checking agents score the share of claims that are hallucinated. Kimi K3.1 Pro 8.30%
- RAGTruth In use A word-level annotated corpus of hallucinations in retrieval-augmented generation answers. It is used to train and evaluate hallucination detection and RAG faithfulness. Nemotron 3.5 Bolt E4B 77.9%