RRepoGEO

REPOGEO REPORT · LITE

THUDM/LongBench

Default branch main · commit 2e00731f · scanned 5/26/2026, 9:38:01 PM

GitHub: 1,178 stars · 132 forks

AI VISIBILITY SCORE
73 /100
Needs work
Category recall
1 / 2
Avg rank #1.0 when recommended
Rule findings
2 pass · 0 warn · 0 fail
Objective metadata checks
AI knows your name
3 / 3
Direct prompts that named your repo
HOW TO READ THIS REPORT

Action plan is what to do next — copy-pasteable changes prioritized by impact. Category visibility is the real GEO test: when a user asks an AI a brand-free question that should surface THUDM/LongBench, does the AI actually recommend you — or your competitors? Objective checks verify the metadata signals AI engines weight first. Self-mention check detects whether AI even knows you exist by name.

Action plan — copy-paste fixes

3 prioritized changes generated by gemini-2.5-flash. Mark items done after you ship the fix.

OVERALL DIRECTION
  • highreadme#1
    Strengthen the README's opening sentence to explicitly position LongBench as an LLM performance evaluation benchmark.

    Why:

    CURRENT
    LongBench v2 is designed to assess the ability of LLMs to handle long-context problems requiring **deep understanding and reasoning** across real-world multitasks.
    COPY-PASTE FIX
    LongBench v2 is a comprehensive benchmark designed for the rigorous evaluation of Large Language Model (LLM) performance on long-context problems, specifically focusing on deep understanding and reasoning across real-world multitasks.
  • mediumtopics#2
    Add more specific topics to improve categorization for performance assessment and document comprehension.

    Why:

    CURRENT
    benchmark, llm, long-context, longtext
    COPY-PASTE FIX
    benchmark, llm, long-context, longtext, llm-evaluation, document-comprehension, reasoning, understanding
  • lowabout#3
    Expand the repository description to clearly state its purpose as an LLM benchmark.

    Why:

    CURRENT
    LongBench v2 and LongBench (ACL 25'&24')
    COPY-PASTE FIX
    LongBench is a comprehensive benchmark for evaluating Large Language Models (LLMs) on their ability to handle long-context problems, focusing on deep understanding and reasoning across realistic multitasks.

Category GEO backends resolved for this scan: google/gemini-2.5-flash, deepseek/deepseek-v4-flash

Category visibility — the real GEO test

Brand-free queries asked to google/gemini-2.5-flash. Did AI recommend you, or someone else?

Same questions for every model — switch tabs to compare answers and rankings.

Recall
1 / 2
50% of queries surface THUDM/LongBench
Avg rank
#1.0
Lower is better. #1 = top recommendation.
Share of voice
4%
Of all named tools, what % are you?
Top rival
L-Eval
Recommended in 1 of 2 queries
COMPETITOR LEADERBOARD
  1. L-Eval · recommended 1×
  2. Needle-in-a-Haystack (NIAH) Test · recommended 1×
  3. OpenAI API · recommended 1×
  4. Anthropic Claude API · recommended 1×
  5. Google Gemini API · recommended 1×
  • CATEGORY QUERY
    How can I effectively benchmark large language models for extended context understanding?
    you: #1
    AI recommended (in order):
    1. LongBench ← you
    2. L-Eval
    3. Needle-in-a-Haystack (NIAH) Test
    4. OpenAI API
    5. Anthropic Claude API
    6. Google Gemini API
    7. RAGAS
    8. LlamaIndex's evaluation modules
    9. Hugging Face `transformers`
    10. GPT-2
    11. LLaMA
    12. Mistral
    13. CNN/DailyMail
    14. XSum
    15. LongSumm
    16. ROUGE
    17. BERTScore
    18. Hugging Face `evaluate`
    19. HumanEval
    20. MBPP
    Show full AI answer
  • CATEGORY QUERY
    What tools help assess LLM performance on complex, very long document comprehension tasks?
    you: not recommended
    AI recommended (in order):
    1. LlamaIndex
    2. LangChain
    3. Ragas
    4. DeepEval
    5. Humanloop
    6. Weights & Biases (W&B) Prompts
    7. Galileo

    AI recommended 7 alternatives but never named THUDM/LongBench. This is the gap to close.

    Show full AI answer

Objective checks

Rule-based audits of metadata signals AI engines weight most.

  • Metadata completeness
    pass

  • README presence
    pass

Self-mention check

Does AI even know your repo exists when asked about it directly?

  • Compared to common alternatives in this category, what is the core differentiator of THUDM/LongBench?
    pass
    AI named THUDM/LongBench explicitly

    AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

  • If a team adopts THUDM/LongBench in production, what risks or prerequisites should they evaluate first?
    pass
    AI named THUDM/LongBench explicitly

    AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

  • In one sentence, what problem does the repo THUDM/LongBench solve, and who is the primary audience?
    pass
    AI named THUDM/LongBench explicitly

    AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

Embed your GEO score

Drop this badge into the README of THUDM/LongBench. It auto-updates whenever the report is rescanned and links back to the latest report — easy public proof that you care about AI discoverability.

RepoGEO badge previewLive preview
MARKDOWN (README)
[![RepoGEO](https://repogeo.com/badge/THUDM/LongBench.svg)](https://repogeo.com/en/r/THUDM/LongBench)
HTML
<a href="https://repogeo.com/en/r/THUDM/LongBench"><img src="https://repogeo.com/badge/THUDM/LongBench.svg" alt="RepoGEO" /></a>
Pro

Subscribe to Pro for deep diagnoses

THUDM/LongBench — Lite scans stay free; this card itemizes Pro deep limits vs Lite.

  • Deep reports10 / month
  • Brand-free category queries5 vs 2 in Lite
  • Prioritized action items8 vs 3 in Lite