https://themodelwire.com/article/minimax-to-release-2-7-trillion-parameter-model-as-open-source-01KX0WD9N3GY9AEVADYHN286RW Modelwire en 2026-07-08T12:44:52.000Z MiniMax to release 2.7 trillion parameter model as open source https://themodelwire.com/article/meta-advances-always-on-lifelogging-glasses-with-embedded-ai-processing-01KX0RZD5A5NYM26ZJZRVVVSXD Modelwire en 2026-07-08T11:48:30.000Z Meta advances always-on lifelogging glasses with embedded AI processing https://themodelwire.com/article/metas-muse-image-agent-faces-eu-legal-collision-over-instagram-photo-use-01KX0Q8FXTTD0T3TTRX37KXNWX Modelwire en 2026-07-08T11:16:37.000Z Meta's Muse Image agent faces EU legal collision over Instagram photo use https://themodelwire.com/article/reasoning-llms-face-new-denial-of-service-attack-vector-01KX0PD0GXY2JERE4REHFBZ66C Modelwire en 2026-07-08T11:00:01.000Z Reasoning LLMs face new denial-of-service attack vector https://themodelwire.com/article/former-deepmind-exec-warns-us-nationalism-in-ai-risks-triggering-arms-race-colla-01KX0J3MFBR26DRPAPFMZN3J35 Modelwire en 2026-07-08T09:30:00.000Z Former DeepMind exec warns US nationalism in AI risks triggering arms race collapse https://themodelwire.com/article/openai-releases-gpt-5-6-after-u-s-government-testing-mandate-01KX0C3DHZCYZZ2TSPK5N74XCK Modelwire en 2026-07-08T08:00:30.000Z OpenAI releases GPT-5.6 after U.S. government testing mandate https://themodelwire.com/article/zml-open-sources-cross-chip-inference-optimizer-backed-by-lecun-01KX0C3BA13GZXP5SQ5SJB4ASV Modelwire en 2026-07-08T08:00:00.000Z ZML open-sources cross-chip inference optimizer backed by LeCun https://themodelwire.com/article/sambanova-hits-11b-valuation-on-back-to-back-mega-rounds-01KX09GZYACD0Z7R62CCQ0YW9J Modelwire en 2026-07-08T07:16:00.000Z SambaNova hits $11B valuation on back-to-back mega-rounds https://themodelwire.com/article/mufg-deploys-chatgpt-enterprise-to-35-000-employees-01KWZHFZNAJ1FC75X0SY9PS9KR Modelwire en 2026-07-08T00:02:05.000Z MUFG deploys ChatGPT Enterprise to 35,000 employees https://themodelwire.com/article/raycast-brings-llm-powered-app-building-to-the-desktop-01KWZGMG1Q5454C2HPEW5E0Y54 Modelwire en 2026-07-08T00:00:29.000Z Raycast brings LLM-powered app building to the desktop https://themodelwire.com/article/meta-launches-muse-image-generator-targeting-advertising-and-creator-workflows-01KWZAM6ZTK3VVSVG6ZMVN7CED Modelwire en 2026-07-07T22:18:10.000Z Meta launches Muse image generator targeting advertising and creator workflows https://themodelwire.com/article/meta-makes-instagram-photos-opt-out-for-muse-image-training-01KWZ9RR6TQZ8KFCJVN9GADEGT Modelwire en 2026-07-07T21:59:29.000Z Meta makes Instagram photos opt-out for Muse Image training https://themodelwire.com/article/openais-safety-research-lead-joshua-achiam-departs-after-nine-years-01KWZ81SXM0M7NE1KABB2NCVDZ Modelwire en 2026-07-07T21:30:00.000Z OpenAI's safety research lead Joshua Achiam departs after nine years https://themodelwire.com/article/hugging-face-models-now-deploy-directly-from-sagemaker-studio-01KWZ768ZTCE980P8TSE84PKV2 Modelwire en 2026-07-07T21:15:33.000Z Hugging Face models now deploy directly from SageMaker Studio https://themodelwire.com/article/meta-deploys-muse-image-model-across-instagram-and-whatsapp-01KWZ4KZ4Y4HKHWRRQZK64JNF1 Modelwire en 2026-07-07T20:31:58.000Z Meta deploys Muse Image model across Instagram and WhatsApp https://themodelwire.com/article/frontier-labs-and-open-source-occupy-separate-ai-market-phases-01KWZ3REVDESNK0K4SF85WJ27H Modelwire en 2026-07-07T20:04:32.000Z Frontier labs and open source occupy separate AI market phases https://themodelwire.com/article/microsoft-builds-internal-ai-models-to-cut-third-party-spending-01KWZ2X1E9MEJXVGDDRMKZ657V Modelwire en 2026-07-07T19:58:20.000Z Microsoft builds internal AI models to cut third-party spending https://themodelwire.com/article/discords-ai-moderation-system-misfires-suspends-8-000-users-over-harmless-images-01KWZ1626RQ5JM9PAM973SN0Q8 Modelwire en 2026-07-07T19:28:38.000Z Discord's AI moderation system misfires, suspends 8,000 users over harmless images https://themodelwire.com/article/microsoft-replaces-openai-and-anthropic-models-with-internal-mai-stack-01KWYYKR2TD33QC5DCVDK0Q0DH Modelwire en 2026-07-07T18:35:58.000Z Microsoft replaces OpenAI and Anthropic models with internal MAI stack https://themodelwire.com/article/new-method-preserves-3d-geometry-detail-in-multimodal-foundation-models-01KWZXGG2X97MQB2T360974WVW Modelwire en 2026-07-07T17:59:50.000Z New method preserves 3D geometry detail in multimodal foundation models https://themodelwire.com/article/cohere-releases-open-source-arabic-speech-model-outperforming-whisper-01KWYW1BY7A67EFA7F5WY8EP19 Modelwire en 2026-07-07T17:54:08.000Z Cohere releases open-source Arabic speech model outperforming Whisper https://themodelwire.com/article/linear-attention-fails-at-graph-denoising-researchers-propose-spectral-alternati-01KWZXGG4181V24XXAHCA01P0P Modelwire en 2026-07-07T17:52:17.000Z Linear attention fails at graph denoising, researchers propose spectral alternative https://themodelwire.com/article/indic-nlp-research-maps-cultural-costs-of-language-ai-homogenization-01KWZYBYWAKVJJY86XVBJQ3SVB Modelwire en 2026-07-07T17:51:10.000Z Indic NLP research maps cultural costs of language AI homogenization https://themodelwire.com/article/unsupervised-parsing-methods-now-applicable-to-animal-communication-systems-01KWZYBYXCFE5JWYF6YTE01NMX Modelwire en 2026-07-07T17:47:26.000Z Unsupervised parsing methods now applicable to animal communication systems https://themodelwire.com/article/anthropic-expands-claude-cowork-to-mobile-and-web-platforms-01KWYV5T8C2YEN7GGS1P4RRW21 Modelwire en 2026-07-07T17:46:59.000Z Anthropic expands Claude Cowork to mobile and web platforms https://themodelwire.com/article/gradient-conflicts-expose-full-duplex-speech-model-bottleneck-01KWZYBYYK51DQ4N8XS07NNW6C Modelwire en 2026-07-07T17:43:36.000Z Gradient conflicts expose full-duplex speech model bottleneck https://themodelwire.com/article/graphbu-generates-realistic-milp-instances-for-solver-training-01KWZXGG58XT4QXX1TTJ2MHKW7 Modelwire en 2026-07-07T17:39:06.000Z GraphBU generates realistic MILP instances for solver training https://themodelwire.com/article/decoupled-ai-framework-lets-hospitals-swap-cancer-models-without-rebuilding-pipe-01KWZXGG6213A5AR9X41X06CCE Modelwire en 2026-07-07T17:38:47.000Z Decoupled AI framework lets hospitals swap cancer models without rebuilding pipelines https://themodelwire.com/article/satellite-imagery-and-morphology-map-urban-poverty-across-indian-cities-01KWZYBYZD3VNK2HM183Q7S5J1 Modelwire en 2026-07-07T17:37:03.000Z Satellite imagery and morphology map urban poverty across Indian cities https://themodelwire.com/article/differentiable-graph-reasoning-bridges-semantic-gaps-in-knowledge-qa-01KWZYBYZV3KPGF218VG06XYN2 Modelwire en 2026-07-07T17:32:36.000Z Differentiable graph reasoning bridges semantic gaps in knowledge QA https://themodelwire.com/article/tpg-deploys-chatgpt-to-compress-investment-research-cycles-01KWYTADV808SGSDP9178DZNPY Modelwire en 2026-07-07T17:15:23.000Z TPG deploys ChatGPT to compress investment research cycles https://themodelwire.com/article/dynakrag-learns-adaptive-control-for-multi-hop-retrieval-augmented-generation-01KWZYBZ08A0GPQ80AZPXG81Y4 Modelwire en 2026-07-07T17:09:36.000Z DynaKRAG learns adaptive control for multi-hop retrieval augmented generation https://themodelwire.com/article/anthropic-brings-claude-cowork-agent-to-mobile-and-web-platforms-01KWYSEZWP8CQSH0XQAR6X2PMG Modelwire en 2026-07-07T17:04:30.000Z Anthropic brings Claude Cowork agent to mobile and web platforms https://themodelwire.com/article/entropath-uses-path-ensembles-to-recover-true-geodesic-structure-in-graph-embedd-01KWZXGG6QYC6VJPPR1E4Y6S5C Modelwire en 2026-07-07T16:58:00.000Z EntroPath uses path ensembles to recover true geodesic structure in graph embeddings https://themodelwire.com/article/pitwall-builds-f1-briefings-with-real-time-claim-verification-gates-01KWZXGG7VRSQ9FSZEG2CJYFMC Modelwire en 2026-07-07T16:55:39.000Z Pitwall builds F1 briefings with real-time claim verification gates https://themodelwire.com/article/data-augmentation-fails-to-block-poisoning-attacks-on-autonomous-vehicle-dataset-01KWZXGG93FHSV7Q0GKX3ER1V9 Modelwire en 2026-07-07T16:46:16.000Z Data augmentation fails to block poisoning attacks on autonomous vehicle datasets https://themodelwire.com/article/new-benchmark-exposes-llm-gaps-in-real-world-data-analysis-01KWZYBZ0T1QCT4G9PTQX4HGXX Modelwire en 2026-07-07T16:43:05.000Z New benchmark exposes LLM gaps in real-world data analysis https://themodelwire.com/article/deepseek-publishes-inference-speed-optimization-technique-01KWYRKEX5XJ4HVG45Z4E2M460 Modelwire en 2026-07-07T16:33:29.000Z DeepSeek publishes inference speed optimization technique https://themodelwire.com/article/quantum-machine-learning-shows-provable-separation-for-many-body-dynamics-01KWZXGG9P3RVKQV4VNVS6SYFR Modelwire en 2026-07-07T16:30:09.000Z Quantum machine learning shows provable separation for many-body dynamics https://themodelwire.com/article/anthropic-brings-claude-cowork-to-mobile-and-web-for-max-subscribers-01KWYPWFEZGFA18T70D1FBBV91 Modelwire en 2026-07-07T16:27:18.000Z Anthropic brings Claude Cowork to mobile and web for Max subscribers https://themodelwire.com/article/wordvoice-enables-word-level-acoustic-control-for-llm-based-speech-synthesis-01KWZYBZ1B9GN17HTC8T143Z5Q Modelwire en 2026-07-07T16:22:59.000Z WordVoice enables word-level acoustic control for LLM-based speech synthesis https://themodelwire.com/article/gpt-5-5-generates-working-web-component-from-conversational-prompt-01KWZ9RQ4T3GSSXJWF9NFPNJ6F Modelwire en 2026-07-07T16:18:16.000Z GPT-5.5 generates working web component from conversational prompt https://themodelwire.com/article/china-prepares-ai-model-export-restrictions-forcing-europe-to-choose-sides-01KWYP1479BNEGZ2M31EY2J12D Modelwire en 2026-07-07T16:13:45.000Z China prepares AI model export restrictions, forcing Europe to choose sides https://themodelwire.com/article/biomedical-qa-framework-routes-llm-inference-by-question-type-01KWZNS9J18EDNRMDCX83BZ53E Modelwire en 2026-07-07T16:12:51.000Z Biomedical QA framework routes LLM inference by question type https://themodelwire.com/article/danus-coordinates-parallel-mathematical-reasoning-agents-via-shared-fact-graph-01KWZNS9K8DZRBVB3A5JEQB8VH Modelwire en 2026-07-07T16:11:30.000Z Danus coordinates parallel mathematical reasoning agents via shared fact graph https://themodelwire.com/article/waymo-vehicles-reported-teens-to-police-raising-autonomy-surveillance-questions-01KWYP130A5AAZG6Q8M70GP30Q Modelwire en 2026-07-07T16:06:31.000Z Waymo vehicles reported teens to police, raising autonomy surveillance questions https://themodelwire.com/article/anthropic-brings-claude-cowork-agent-to-smartphones-with-background-task-persist-01KWYN5K2KR017WR9DW7YPC7SS Modelwire en 2026-07-07T16:00:00.000Z Anthropic brings Claude Cowork agent to smartphones with background task persistence https://themodelwire.com/article/new-technique-balances-concept-removal-with-model-quality-in-diffusion-systems-01KWZNS9J831KSZTASMMDDFGXQ Modelwire en 2026-07-07T15:59:59.000Z New technique balances concept removal with model quality in diffusion systems https://themodelwire.com/article/rubench-benchmarks-coding-agents-on-native-russian-task-specifications-01KWZNS9KWWAYH8FM2R4XA9PWZ Modelwire en 2026-07-07T15:41:22.000Z RuBench benchmarks coding agents on native Russian task specifications https://themodelwire.com/article/language-bridges-vision-and-smell-in-multimodal-ai-framework-01KWZNS9KAKZNAZCRZV315SY8E Modelwire en 2026-07-07T15:31:55.000Z Language bridges vision and smell in multimodal AI framework