https://ai-tldr.dev/releases/anthropic-claude-code-gitlab-support/ AI/TLDR en 2026-08-14T12:00:00+00:00 Claude Code 2.1.233 — GitLab merge requests, marketplaces and token redaction https://ai-tldr.dev/releases/google-gemini-media-watermark-toggle/ AI/TLDR en 2026-08-14T12:00:00+00:00 Gemini watermarks become optional — Google adds an off switch for AI media https://ai-tldr.dev/releases/openai-computer-history/ AI/TLDR en 2026-08-13T12:00:00+00:00 Computer History — ChatGPT builds a memory from your Mac activity https://ai-tldr.dev/releases/anthropic-claude-code-session-costs/ AI/TLDR en 2026-08-14T12:00:00+00:00 Claude Code token costs — Anthropic explains what makes sessions expensive https://ai-tldr.dev/releases/mixedbread-toast-1/ AI/TLDR en 2026-08-13T12:00:00+00:00 Toast 1 — Mixedbread's search model runs the whole retrieval loop https://ai-tldr.dev/releases/1littlecoder-gemini-3-7-flash-cost-aug14/ AI/TLDR en 2026-08-14T12:00:00+00:00 1littlecoder — 'Save your token cost with Gemini 3.7 Flash' https://ai-tldr.dev/releases/google-credentio/ AI/TLDR en 2026-08-13T12:00:00+00:00 Credentio — Google open-sources the C++ library behind its content credentials https://ai-tldr.dev/releases/google-heir-encrypted-inference/ AI/TLDR en 2026-08-14T12:00:00+00:00 HEIR — Google's compiler runs AI models on encrypted data https://ai-tldr.dev/releases/fireship-flock-edge-ml-aug14/ AI/TLDR en 2026-08-14T12:00:00+00:00 Fireship — 'The edge ML pipeline that jailbroke the 4th Amendment' https://ai-tldr.dev/releases/qwen-3-8-27b/ AI/TLDR en 2026-08-14T12:00:00+00:00 Qwen3.8-27B — a 27B open model that beats Opus 4.6 Max on SWE-bench Pro https://ai-tldr.dev/releases/suno-studio-2-0/ AI/TLDR en 2026-08-13T12:00:00+00:00 Suno Studio 2.0 — browser music workstation adds MIDI and a chat bar https://ai-tldr.dev/releases/mun-logadan-opus-5-feels-worse-aug14/ AI/TLDR en 2026-08-14T12:00:00+00:00 Mun logadan — benchmarks reward guessing, so Claude Opus 5 stops asking https://ai-tldr.dev/releases/zai-glm-5-3/ AI/TLDR en 2026-08-14T12:00:00+00:00 GLM-5.3 — Z.ai's coding model improves without retraining the base https://ai-tldr.dev/releases/two-minute-papers-claude-riemann-aug14/ AI/TLDR en 2026-08-14T12:00:00+00:00 Two Minute Papers — 'Claude AI Failed 650 Times, Then Beat The Human Record' https://ai-tldr.dev/releases/huggingface-icml-2026-reproductions/ AI/TLDR en 2026-08-13T12:00:00+00:00 ICML 2026 Open Reproductions — agents re-ran 2,226 papers, contested 496 https://ai-tldr.dev/releases/cursor-cloud-agent-builds/ AI/TLDR en 2026-08-13T12:00:00+00:00 Cursor Builds — cloud agents fork a warm dev environment instead of setup https://ai-tldr.dev/releases/wes-roth-grok-4-6-fable-aug13/ AI/TLDR en 2026-08-13T12:00:00+00:00 Wes Roth — 'Grok 4.6 is Fable now' https://ai-tldr.dev/releases/minimax-music-3/ AI/TLDR en 2026-08-13T12:00:00+00:00 MiniMax Music 3.0 — open-weights model writes a full five-minute song https://ai-tldr.dev/releases/writer-palmyra-x6/ AI/TLDR en 2026-08-13T12:00:00+00:00 Palmyra X6 — Writer's flagship model halves the cost of an agent task