mlsanityData to 14 Oct 2026

Flash-VL 2B: Optimizing Vision-Language Model Performance for Ultra-Low Latency and High Throughput

arXiv 2505.09498 ↗First posted 14 May 2025cs.CV cs.AIcited by 4

On the map: Multimodal LLMs / vision-language models → Visual token pruning and efficiency

Abstract (arXiv)

In this paper, we introduce Flash-VL 2B, a novel approach to optimizing Vision-Language Models (VLMs) for real-time applications, targeting ultra-low latency and high throughput without sacrificing accuracy. Leveraging advanced architectural enhancements and efficient computational strategies, Flash-VL 2B is designed to maximize throughput by reducing processing time while maintaining competitive performance across multiple vision-language benchmarks. Our approach includes tailored architectural choices, token compression mechanisms, data curation, training schemes, and a novel image processing technique called implicit semantic stitching that effectively balances computational load and model performance. Through extensive evaluations on 11 standard VLM benchmarks, we demonstrate that Flash-VL 2B achieves state-of-the-art results in both speed and accuracy, making it a promising solution for deployment in resource-constrained environments and large-scale real-time applications.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Placed in Visual token pruning and efficiency, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 77 of its 80 references that are in the map. First posted on arXiv 14 May 2025.

Builds on

77 of 80 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024FastVLM: Efficient Vision Encoding for Vision Language Models17 Dec
  2. 2023MobileVLM : A Fast, Strong and Open Vision Language Assistant for Mobile Devices28 Dec

Other references · 75

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2024The Llama 3 Herd of Modelscited by 14,715
  4. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  5. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  6. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  7. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  8. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  9. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  10. 2024Qwen2.5 Technical Reportcited by 4,942
  11. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,703
  12. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
38 more
  1. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
  2. 2023Qwen Technical Reportcited by 3,344
  3. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,109
  4. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,719
  5. 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,565
  6. 2023Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyondcited by 2,224
  7. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,141
  8. 2023MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGIcited by 2,080
  9. 2024Gemma 2: Improving Open Language Models at a Practical Sizecited by 1,998
  10. 2024Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phonecited by 1,917
  11. 2025Gemma 3 Technical Reportcited by 1,824
  12. 2020GLU Variants Improve Transformercited by 1,804
  13. 2019Towards VQA Models That Can Readcited by 1,718
  14. 2023MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Modelscited by 1,652
  15. 2023MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contextscited by 1,614
  16. 2019Root Mean Square Layer Normalizationcited by 1,480
  17. 2022ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoningcited by 1,479
  18. 2020DocVQA: A Dataset for VQA on Document Imagescited by 1,317
  19. 2023MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilitiescited by 1,239
  20. 2025SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Featurescited by 1,188
  21. 2024How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suitescited by 1,050
  22. 2023ShareGPT4V: Improving Large Multi-Modal Models with Better Captionscited by 1,005
  23. 2023SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehensioncited by 940
  24. 2024Are We on the Right Way for Evaluating Large Vision-Language Models?cited by 914
  25. 2016A Diagram Is Worth A Dozen Imagescited by 804
  26. 2023EVA-CLIP: Improved Training Techniques for CLIP at Scalecited by 798
  27. 2024MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategiescited by 723
  28. 2023LLM-Pruner: On the Structural Pruning of Large Language Modelscited by 650
  29. 2023DataComp: In search of the next generation of multimodal datasetscited by 644
  30. 2024MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?cited by 586
  31. 2024LLaVA-NeXT-Interleave: Tackling Multi-image, Video, and 3D in Large Multimodal Modelscited by 584
  32. 2023HallusionBench: An Advanced Diagnostic Suite for Entangled Language Hallucination and Visual Illusion in Large Vision-Language Modelscited by 583
  33. 2023Textbooks Are All You Needcited by 569
  34. 2023Textbooks Are All You Need II: phi-1.5 technical reportcited by 565
  35. 2024Measuring Multimodal Mathematical Reasoning with MATH-Vision Datasetcited by 526
  36. 2025Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAscited by 521
  37. 2015Compositional Semantic Parsing on Semi-Structured Tablescited by 515
  38. 2024MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmarkcited by 512

50 most cited of 75 All 75 (JSON)

Built on by

Cited by year

Citing papers per year: 2025: 4.

4 papers.

  1. 2025MiniCPM-V 4.5: Cooking Efficient MLLMs via Architecture, Data, and Training Recipecited by 186
  2. 2025Affordance Benchmark for MLLMscited by 5
  3. 2025Less Redundancy: Boosting Practicality of Vision Language Model in Walking Assistantscited by 2
  4. 2025CLGRPO: Reasoning Ability Enhancement for Small VLMscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Empirical Recipes for Efficient and Compact Vision-Language Models
  2. 2025A Survey on Efficient Vision-Language Models
  3. 2025EffiVLM-BENCH: A Comprehensive Benchmark for Evaluating Training-Free Acceleration in Large Vision-Language Models
  4. 2024VL-Cache: Sparsity and Modality-Aware KV Cache Compression for Vision-Language Model Inference Acceleration
  5. 2025Benchmarking and Enhancing VLM for Compressed Image Understanding
  6. 2026UVLM: A Universal Vision-Language Model Loader for Reproducible Multimodal Benchmarking
  7. 2024VLsI: Verbalized Layers-to-Interactions from Large to Small Vision Language Models
  8. 2025MagicVL-2B: Empowering Vision-Language Models on Mobile Devices with Lightweight Visual Encoders via Curriculum Learning