meddown
OSV published a cluster of 13 advisories (PYSEC-2026-1977 through PYSEC-2026-1988) affecting the Hugging Face...
OSV published a cluster of 13 advisories (PYSEC-2026-1977 through PYSEC-2026-1988) affecting the Hugging Face transformers library on 2026-07-07, the latest security signal as of 2026-08-03.
window 14devidence 87confidence score 100
confidence score
Strong evidence: 2 independent source classes support this read.
100
medium confidence2 independent source classesotherpasses publish gate
signal brief
OSV published a cluster of 13 advisories (PYSEC-2026-1977 through PYSEC-2026-1988) affecting the Hugging Face transformers library on 2026-07-07, the latest security signal as of 2026-08-03. The cluster spans three bug classes: regular expression denial of service (ReDoS) in DonutProcessor, SETTING_RE, AdamWeightDecay optimizer, MarianTokenizer, get_imports(), and get_configuration_file; deserialization of untrusted data; and username injection via improper input validation. Advisories: PYSEC-2026-1977, PYSEC-2026-1978, PYSEC-2026-1979, PYSEC-2026-1980, PYSEC-2026-1981, PYSEC-2026-1982, PYSEC-2026-1983, PYSEC-2026-1984, PYSEC-2026-1985, PYSEC-2026-1986, PYSEC-2026-1987, PYSEC-2026-1988.
This is a materially negative dev-trust signal for Hugging Face. transformers is foundational AI infrastructure, and a concentrated batch of parsing/deserialization CVEs can force downstream enterprise teams to halt upgrades, add mitigations, and re-evaluate trust in the ecosystem. The contemporaneous model/dataset upload activity in the same window — mostly zero-download personal repositories and unrelated datasets — is noise and does not offset the vulnerability cluster.
What the sources said
- "Transformers is vulnerable to ReDoS attack through its DonutProcessor class" — PYSEC-2026-1977
- "Transformers Deserialization of Untrusted Data vulnerability" — PYSEC-2026-1978
- "Hugging Face Transformers vulnerable to Regular Expression Denial of Service (ReDoS) in the AdamWeightDecay optimizer" — PYSEC-2026-1980
- "Transformers's Improper Input Validation vulnerability can be exploited through username injection" — PYSEC-2026-1986
source data used
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: transformers, safetensors, generated_from_trainer, unsloth, sft, trl, base_model:unsloth/Qwen3-8B, base_model:finetune:unsloth/Qwen3-8B, endpoints_compatible, region:us”
“Downloads: 0 Likes: 4 Tags: region:us”
“Downloads: 494 Likes: 0 Tags: transformers, gguf, merlina, grimoire, image-text-to-text, vision-language-model, orpo, en, dataset:nbeerbower/weasel-dpo, dataset:nbeerbower/seX-ai-dpo, dataset:nbeerbower/grok-politically-incorrect-dpo, base_model:nbeerbower/CHUD-Qwen3.6-27B”
“Downloads: 0 Likes: 0 Tags: safetensors, gpt2, region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: transformers, safetensors, generated_from_trainer, sft, trl, unsloth, base_model:unsloth/Llama-3.2-3B-Instruct-bnb-4bit, base_model:finetune:unsloth/Llama-3.2-3B-Instruct-bnb-4bit, endpoints_compatible, region:us”
“Downloads: 0 Likes: 0 Tags: transformers, gguf, laguna, moe, uncensored, abliterix, safetensors, bf16, poolside, en, base_model:SC117/Laguna-S-2.1-Uncensored, base_model:quantized:SC117/Laguna-S-2.1-Uncensored”
“Downloads: 37 Likes: 0 Tags: safetensors, xlm-roberta, pos-tagging, token-classification, balochi, arabic-script, nlp, low-resource, bal, license:apache-2.0, region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 365 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, instruction-tuned, custom_code, en, license:apache-2.0, region:us”
“Downloads: 0 Likes: 1 Tags: region:us”
“Downloads: 0 Likes: 1 Tags: region:us”
“Downloads: 898 Likes: 1 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, instruction-tuned, custom_code, en, license:apache-2.0, region:us”
“Downloads: 70 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, instruction-tuned, custom_code, en, license:apache-2.0, region:us”
“Downloads: 17 Likes: 1 Tags: transformers, safetensors, qwen3_5, image-text-to-text, efficient, qwen, qwen3.5, nomi, lazyloopstudio, unsloth, nomi2, conversational”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 78 Likes: 1 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, instruction-tuned, custom_code, en, license:apache-2.0, region:us”
“Downloads: 76 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, instruction-tuned, custom_code, en, license:apache-2.0, region:us”
“Downloads: 318 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, instruction-tuned, custom_code, en, license:apache-2.0, region:us”
“Downloads: 77 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, instruction-tuned, custom_code, en, license:apache-2.0, region:us”
“Downloads: 0 Likes: 7 Tags: region:us”
“Downloads: 68 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, instruction-tuned, custom_code, en, license:apache-2.0, region:us”
“Downloads: 163 Likes: 1 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, instruction-tuned, custom_code, en, license:apache-2.0, region:us”
“Downloads: 438 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, instruction-tuned, custom_code, en, license:apache-2.0, region:us”
“Downloads: 36 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, instruction-tuned, custom_code, en, license:apache-2.0, region:us”
“Downloads: 38 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, instruction-tuned, custom_code, en, license:apache-2.0, region:us”
“Downloads: 553 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, pretrained, custom_code, en, license:apache-2.0, region:us”
“Downloads: 125 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, pretrained, custom_code, en, license:apache-2.0, region:us”
“Downloads: 110 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, pretrained, custom_code, en, license:apache-2.0, region:us”
“Downloads: 130 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, pretrained, custom_code, en, license:apache-2.0, region:us”
“Downloads: 8516 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, pretrained, custom_code, en, license:apache-2.0, region:us”
“Downloads: 117 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, pretrained, custom_code, en, license:apache-2.0, region:us”
“Downloads: 107 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, pretrained, custom_code, en, license:apache-2.0, region:us”
“Downloads: 125 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, pretrained, custom_code, en, license:apache-2.0, region:us”
“Downloads: 107 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, pretrained, custom_code, en, license:apache-2.0, region:us”
“Downloads: 0 Likes: 0 Tags: license:apache-2.0, region:us”
“Downloads: 360 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, pretrained, custom_code, en, license:apache-2.0, region:us”
“Downloads: 0 Likes: 0 Tags: license:apache-2.0, region:us”
“Downloads: 100 Likes: 1 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, pretrained, custom_code, en, license:apache-2.0, region:us”
“Downloads: 0 Likes: 0 Tags: license:mit, region:us”
“Downloads: 0 Likes: 0 Tags: transformers, safetensors, qwen2, text-generation, trl, sft, conversational, arxiv:1910.09700, text-generation-inference, endpoints_compatible, region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 1491 Likes: 0 Tags: safetensors, pit, causal-lm, point-in-time, temporal-llm, pretrained, custom_code, en, license:apache-2.0, region:us”
“Downloads: 0 Likes: 0 Tags: transformers, onnx, marian, text2text-generation, translation, ga, en, base_model:Helsinki-NLP/opus-mt-ga-en, base_model:quantized:Helsinki-NLP/opus-mt-ga-en, license:apache-2.0, endpoints_compatible, region:us”
“Downloads: 0 Likes: 3 Tags: license:unknown, region:us”
“Downloads: 14 Likes: 0 Tags: transformers, safetensors, wav2vec2-bert, automatic-speech-recognition, arxiv:1910.09700, endpoints_compatible, region:us”
“Downloads: 0 Likes: 1 Tags: region:us”
“Downloads: 1521 Likes: 0 Tags: region:us”
“Downloads: 13546 Likes: 16 Tags: task_categories:time-series-forecasting, license:mit, region:us”
“Downloads: 2981 Likes: 3 Tags: task_categories:text-generation, language:en, license:apache-2.0, size_categories:1K<n<10K, doi:10.57967/hf/9695, region:us, token-compression, context-pruning, kompress, kompress-ultra, agent-loops, opencode”
“Downloads: 557 Likes: 0 Tags: region:us”
“Downloads: 2082 Likes: 2 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 781 Likes: 2 Tags: license:mit, region:us”
“Downloads: 7409 Likes: 4 Tags: license:apache-2.0, region:us”
“Downloads: 5458 Likes: 2 Tags: language:en, license:cc-by-4.0, region:us, prediction-markets, polymarket, finance, time-series, order-books”
“Downloads: 628 Likes: 0 Tags: language:hr, region:us, bike-sharing, time-series, raw”
“Downloads: 26590 Likes: 338 Tags: task_categories:text-generation, task_categories:feature-extraction, task_categories:text-classification, task_categories:question-answering, language:en, license:odc-by, size_categories:10M<n<100M, modality:text, region:us, hacker-news, forum, text”
“Downloads: 13885 Likes: 0 Tags: region:us”
“Downloads: 824 Likes: 0 Tags: language:hr, region:us, bike-sharing, time-series, zagreb”
“Downloads: 17576 Likes: 7 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 1416 Likes: 3 Tags: region:us”
“Downloads: 2219 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 8288 Likes: 10 Tags: license:mit, region:us, feedback, append-only, public-ledger”
“Downloads: 17075 Likes: 4 Tags: region:us”
“Downloads: 67 Likes: 0 Tags: region:us”
“Downloads: 0 Likes: 0 Tags: region:us”
“Downloads: 698 Likes: 1 Tags: region:us, weibo, social-media, web-archive, multimodal”
“Downloads: 863 Likes: 2 Tags: region:us”
“Downloads: 1900 Likes: 1 Tags: region:us”
“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.