On July 7, 2026, the Open Source Vulnerabilities (OSV) database published 12 new advisories for the Hugging Face...
On July 7, 2026, the Open Source Vulnerabilities (OSV) database published 12 new advisories for the Hugging Face Transformers library, covering multiple Regular Expression Denial of Service (ReDoS) vulnerabilities and a deserialization issue.
confidence score
Strong evidence: 3 independent source classes support this read.
signal brief
On July 7, 2026, the Open Source Vulnerabilities (OSV) database published 12 new advisories for the Hugging Face Transformers library, covering multiple Regular Expression Denial of Service (ReDoS) vulnerabilities and a deserialization issue. The advisories (PYSEC-2026-1977 through 1988) affect various components including DonutProcessor, AdamWeightDecay optimizer, MarianTokenizer, and others. These vulnerabilities could allow attackers to cause service disruption or potentially execute arbitrary code. The disclosures come shortly after the release of Transformers 5.13.0 on July 3 (source 76). While patches are likely forthcoming, the volume of CVEs raises concerns about code quality and security posture, potentially eroding developer trust in the ecosystem.
What the sources said
- OSV advisory PYSEC-2026-1977: "Transformers is vulnerable to ReDoS attack through its DonutProcessor class."
- OSV advisory PYSEC-2026-1978: "Transformers Deserialization of Untrusted Data vulnerability."
- OSV advisory PYSEC-2026-1980: "Hugging Face Transformers vulnerable to Regular Expression Denial of Service (ReDoS) in the AdamWeightDecay optimizer."
- OSV advisory PYSEC-2026-1984: "Transformers Regular Expression Denial of Service (ReDoS) vulnerability."
source data used
“Downloads: 0 Likes: 1 Tags: pytorch, finance, trading, time-series, transformer, moe, grouped-query-attention, stock-prediction, forex-prediction, license:mit, region:us”
“Downloads: 0 Likes: 0 Tags: transformers, gguf, text-generation-inference, unsloth, mistral, grok, unhinged, unaligned, abliterated, x.ai, dpo, sft”
“Downloads: 0 Likes: 0 Tags: transformers, safetensors, qwen2, text-generation, generated_from_trainer, grpo, trl, conversational, arxiv:2402.03300, base_model:Qwen/Qwen2.5-1.5B-Instruct, base_model:finetune:Qwen/Qwen2.5-1.5B-Instruct, text-generation-inference”
“Downloads: 0 Likes: 5 Tags: onnx, arxiv:2602.03139, base_model:circlestone-labs/Anima, base_model:quantized:circlestone-labs/Anima, license:apache-2.0, region:us”
“Downloads: 0 Likes: 1 Tags: safetensors, region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: safetensors, qwen3_5, region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: transformers, gguf, fine-tuned, lora, sft, auto-sft, en, base_model:theprint/Survivor-v1-2B, base_model:adapter:theprint/Survivor-v1-2B, license:mit, endpoints_compatible, region:us”
“Downloads: 28 Likes: 12 Tags: transformers, safetensors, qwen3_vl, image-text-to-text, multimodal, scientific, protein, rna, dna, molecule, weather, medical-imaging”
“Downloads: 0 Likes: 0 Tags: qwen3_5_moe, region:us”
“Downloads: 0 Likes: 0 Tags: safetensors, llama, from-scratch, legal, finance, license:mit, region:us”
“Downloads: 0 Likes: 5 Tags: region:us”
“Downloads: 76 Likes: 0 Tags: lerobot, safetensors, robotics, pi052, dataset:pepijn223/robocasa_pretrain_human300_v4_annotated5, license:apache-2.0, region:us”
“Downloads: 0 Likes: 0 Tags: coreml, automatic-speech-recognition, ios, swift, onnxruntime, apple-neural-engine, on-device, en, arxiv:2605.28139, license:apache-2.0, region:us”
“Downloads: 0 Likes: 0 Tags: peft, safetensors, base_model:adapter:unsloth/llama-3-8b-Instruct-bnb-4bit, lora, transformers, unsloth, text-generation, conversational, arxiv:1910.09700, base_model:unsloth/llama-3-8b-Instruct-bnb-4bit, region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: de, license:apache-2.0, region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: wsss, boundary-refinement, pytorch, image-segmentation, license:mit, region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 2 Likes: 0 Tags: license:apache-2.0, region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: license:cc-by-4.0, region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: transformers, safetensors, generated_from_trainer, sft, trl, base_model:meta-llama/Llama-3.2-3B-Instruct, base_model:finetune:meta-llama/Llama-3.2-3B-Instruct, endpoints_compatible, region:us”
“Downloads: 0 Likes: 0 Tags: license:cc-by-4.0, region:us”
“Downloads: 0 Likes: 0 Tags: peft, safetensors, lora, unsloth, qwen, document-type-classification, apprentice, paperless-gpt, dataset:anirudh1112/corrected-tobacco-dataset-with-ocr, base_model:Qwen/Qwen3.5-4B, base_model:adapter:Qwen/Qwen3.5-4B, license:apache-2.0”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: document-ai, ocr, mineru, jax-js, webgpu, safetensors, image-to-text, license:apache-2.0, region:us”
“Downloads: 0 Likes: 0 Tags: mlx, safetensors, llama, obliteratus, abliteration, uncensored, obliterate, mlx-my-repo, en, base_model:usermma/AMALIA-9B-0626-SFT-Heretic-Abliterated, base_model:finetune:usermma/AMALIA-9B-0626-SFT-Heretic-Abliterated, region:us”
“Downloads: 0 Likes: 0 Tags: transformers, safetensors, generated_from_trainer, trl, sft, base_model:meta-llama/Llama-3.2-3B-Instruct, base_model:finetune:meta-llama/Llama-3.2-3B-Instruct, endpoints_compatible, region:us”
“Downloads: 0 Likes: 0 Tags: pytorch, bert, propaganda, en, endpoints_compatible, region:us”
“Downloads: 0 Likes: 9 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: peft, audio-text-to-text, audio-reasoning, automatic-speech-recognition, arabic, qwen3-omni, moe, lora, ar, en, license:other, region:us”
“Downloads: 0 Likes: 0 Tags: license:cc-by-nc-4.0, region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: safetensors, robotics, vla, vision-language-action, libero, model-compression, arxiv:2606.27755, license:apache-2.0, region:us”
“Downloads: 0 Likes: 0 Tags: mlx, safetensors, qwen2, obliteratus, abliteration, uncensored, obliterate, mlx-my-repo, en, base_model:usermma/DeepHat-V1-7B-Heretic-Abliterated, base_model:quantized:usermma/DeepHat-V1-7B-Heretic-Abliterated, 8-bit”
“Downloads: 0 Likes: 0 Tags: safetensors, region:us”
“Downloads: 0 Likes: 0 Tags: quantized, int4, moe, quark, amd, en, base_model:Qwen/Qwen3.5-35B-A3B, base_model:finetune:Qwen/Qwen3.5-35B-A3B, license:apache-2.0, region:us”
“Downloads: 0 Likes: 0 Tags: transformers, safetensors, qwen3, text-generation, ronpo, mnpo, preference-optimization, conversational, base_model:Qwen/Qwen3-8B, base_model:finetune:Qwen/Qwen3-8B, text-generation-inference, endpoints_compatible”
“Downloads: 0 Likes: 0 Tags: safetensors, qwen3_5_moe, region:us”
“Downloads: 19192 Likes: 4 Tags: region:us”
“Downloads: 64 Likes: 0 Tags: task_categories:image-to-3d, license:cc-by-nc-nd-4.0, size_categories:1K<n<10K, region:us, human, avatar, SMPL-X, FLAME, TED”
“Downloads: 0 Likes: 0 Tags: size_categories:n<1K, format:audiofolder, modality:audio, library:datasets, library:mlcroissant, region:us”
“Downloads: 0 Likes: 4 Tags: license:other, arxiv:2603.02413, region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 17094 Likes: 1 Tags: license:other, size_categories:100K<n<1M, modality:text, region:us, osint, adsb, ais, aviation, maritime, szl-holdings, doi:10.5281/zenodo.19944926”
“Downloads: 1449 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 18756 Likes: 5 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 4085 Likes: 3 Tags: region:us”
“Downloads: 2748 Likes: 1 Tags: region:us”
“Downloads: 6332 Likes: 1 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 100 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 235 Likes: 0 Tags: region:us”
“Downloads: 1866 Likes: 0 Tags: region:us”
“Downloads: 12961 Likes: 14 Tags: region:us”
“Downloads: 10774 Likes: 14 Tags: license:mit, region:us”
“Downloads: 17233 Likes: 0 Tags: region:us”
“Downloads: 11212 Likes: 11 Tags: region:us”
“Downloads: 556 Likes: 0 Tags: region:us”
“Downloads: 36 Likes: 0 Tags: region:us”
“Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.”
“Aliases: CVE-2025-3933, GHSA-37mw-44qp-f5jm Transformers is vulnerable to ReDoS attack through its DonutProcessor class”
“Aliases: CVE-2024-3568, GHSA-37q5-v5qm-c9v8 Transformers Deserialization of Untrusted Data vulnerability”
“Aliases: CVE-2025-3262, GHSA-489j-g2vx-39wf Transformers vulnerable to ReDoS attack through its SETTING_RE variable”
“Aliases: CVE-2025-6921, GHSA-4w7r-h757-3r74 Hugging Face Transformers vulnerable to Regular Expression Denial of Service (ReDoS) in the AdamWeightDecay optimizer”
“Aliases: CVE-2025-6638, GHSA-59p9-h35m-wg4g Hugging Face Transformers is vulnerable to ReDoS through its MarianTokenizer”
“Aliases: CVE-2024-12720, GHSA-6rvg-6v2m-4j46 Transformers Regular Expression Denial of Service (ReDoS) vulnerability”
“Aliases: CVE-2025-5197, GHSA-9356-575x-2w9m Hugging Face Transformers Regular Expression Denial of Service (ReDoS) vulnerability”
“Aliases: CVE-2025-1194, GHSA-fpwr-67px-3qhx Transformers Regular Expression Denial of Service (ReDoS) vulnerability”
“Aliases: CVE-2025-3264, GHSA-jjph-296x-mrcr Transformers vulnerable to ReDoS attack through its get_imports() function”
“Aliases: CVE-2025-3777, GHSA-phhr-52qp-3mj4 Transformers's Improper Input Validation vulnerability can be exploited through username injection”
“Aliases: CVE-2025-3263, GHSA-q2wp-rjmx-x6x9 Transformers's ReDoS vulnerability in get_configuration_file can lead to catastrophic backtracking”
“Aliases: CVE-2025-6051, GHSA-rcv9-qm8p-9p6j Hugging Face Transformers library has Regular Expression Denial of Service”
Decision support, not stock advice. This signal is research with cited evidence — not a recommendation to buy, sell, or hold any security.