https://ai-tldr.dev/releases/tencent-hy4-preview/ AI/TLDR en 2026-08-28T12:00:00+00:00 Tencent Hy4 preview — 770B open-weights model with a 1M-token context https://ai-tldr.dev/releases/anthropic-claude-code-2-1-248/ AI/TLDR en 2026-08-27T12:00:00+00:00 Claude Code 2.1.248 — a restricted mode that strips the tools that run code https://ai-tldr.dev/releases/two-minute-papers-qwen-flash-next-aug28/ AI/TLDR en 2026-08-28T12:00:00+00:00 Two Minute Papers — 'This Free AI Just Caught The Billion Dollar Giants' https://ai-tldr.dev/releases/ai-explained-agi-2026-swarm-aug27/ AI/TLDR en 2026-08-27T12:00:00+00:00 AI Explained — 'Sam Altman: AGI in 2026' and the METR swarm report https://ai-tldr.dev/releases/embracethered-claude-code-auto-mode-rce/ AI/TLDR en 2026-08-26T12:00:00+00:00 Claude Code auto mode broken — a prompt injection chain that runs real code https://ai-tldr.dev/releases/cohere-parse-v5-0/ AI/TLDR en 2026-08-27T12:00:00+00:00 Cohere Parse — a 2.3B document model that turns PDFs into clean Markdown https://ai-tldr.dev/releases/anthropic-claude-for-scientists-seats/ AI/TLDR en 2026-08-27T12:00:00+00:00 Claude for scientists — Anthropic opens 10,000 free and discounted seats https://ai-tldr.dev/releases/louisabraham-load-bearing-vocabulary/ AI/TLDR en 2026-08-27T12:00:00+00:00 The load-bearing vocabulary of Claude — 461,121 pull requests, clustered https://ai-tldr.dev/releases/anthropic-model-hardware-standard/ AI/TLDR en 2026-08-27T12:00:00+00:00 Model Hardware Standard — Anthropic's spec for agents that run lab equipment https://ai-tldr.dev/releases/google-gemini-omni-1-1-flash/ AI/TLDR en 2026-08-27T12:00:00+00:00 Gemini Omni 1.1 Flash — Google's video model gets keyframes and 4K output https://ai-tldr.dev/releases/deepmind-double-blind-evaluations/ AI/TLDR en 2026-08-27T12:00:00+00:00 Double-blind AI evaluations — DeepMind tests Gemini without seeing the prompts https://ai-tldr.dev/releases/calvinfrenchowen-small-models-arrived/ AI/TLDR en 2026-08-26T12:00:00+00:00 Calvin French-Owen — small models are now cheap enough for consumer apps https://ai-tldr.dev/releases/wes-roth-openai-phaseone-aug27/ AI/TLDR en 2026-08-27T12:00:00+00:00 Wes Roth — 'OpenAI just revealed PHASEONE' on the METR incident report https://ai-tldr.dev/releases/google-gemini-3-5-transcribe/ AI/TLDR en 2026-08-26T12:00:00+00:00 Gemini 3.5 Transcribe — Google's speech model cleans up your ums and ahs https://ai-tldr.dev/releases/vllm-v0-28-0/ AI/TLDR en 2026-08-26T12:00:00+00:00 vLLM v0.28.0 — Kimi K3 gets a full-stack speed pass https://ai-tldr.dev/releases/bytedance-diffusion-opsd/ AI/TLDR en 2026-08-25T12:00:00+00:00 DiffusionOPSD — reward training for diffusion models at 40% fewer GPU-hours https://ai-tldr.dev/releases/openai-hugging-face-incident-report/ AI/TLDR en 2026-08-26T12:00:00+00:00 OpenAI's Hugging Face report — the full account of how a test model broke out