Datasets › MMNeedle
MMNeedle (Multimodal Needle in a Haystack)
We introduce the MultiModal Needle-in-a-haystack (MMNeedle) benchmark, specifically designed to assess the long-context capabilities of MLLMs. Besides multi-image input, we employ image stitching to further increase the input context length, and develop a protocol to automatically generate labels for sub-image level retrieval. Essentially, MMNeedle evaluates MLLMs by stress-testing their capability to locate a target sub-image (needle) within a set of images (haystack) based on textual instructions and descriptions of image contents. This setup necessitates an advanced understanding of extensive visual contexts and effective information retrieval within long-context image inputs.
Benchmarks archive 2025-07-28
All 1 leaderboard whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.
| First row (archive order) | Paper | Code | ||||
|---|---|---|---|---|---|---|
| Long-Context Understanding | MMNeedle | GPT-4o 1 Image, 4*4 Stitching, Exact Accuracy 83 | GPT-4 Technical Report | openai/evals +10 | 12 | Compare |
Papers archive 2025-07-28
9 shown of 9 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 12. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.
| Date | Samples run Syntology | |||
|---|---|---|---|---|
| What matters when building vision-language models? | 0 | 1 | 3 May 2024 | not harvested |
| LLaVA-UHD: an LMM Perceiving Any Aspect Ratio and High-Resolution Images | 1 | 1 | 18 Mar 2024 | not harvested |
| Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context | 1 | 1 | 8 Mar 2024 | not harvested |
| The Claude 3 Model Family: Opus, Sonnet, Haiku | 0 | 1 | 4 Mar 2024 | not harvested |
| Gemini: A Family of Highly Capable Multimodal Models | 1 | 1 | 19 Dec 2023 | not harvested |
| mPLUG-Owl2: Revolutionizing Multi-modal Large Language Model with Modality Collaboration | 2 | 1 | 7 Nov 2023 | ran 1 of 3 samples (2 unverified; 3 pointer-only for licence) |
| CogVLM: Visual Expert for Pretrained Language Models | 4 | 2 | 6 Nov 2023 | not harvested |
| InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning | 4 | 2 | 11 May 2023 | not harvested |
| GPT-4 Technical Report | 11 | 2 | 15 Mar 2023 | ran 2 of 5 samples (3 unverified; 1 pointer-only for licence) |
Dataset loaders archive 2025-07-28
No loader listed in the archive.
Tasks archive 2025-07-28
License archive 2025-07-28
CC BY 4.0
Modalities archive 2025-07-28
Languages archive 2025-07-28
Variants archive 2025-07-28
- MMNeedle
1 variant name, as the archive lists them.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections