Fresh stories
Moonshot launches Kimi K3 with 2.8T parameters and 1M context
Moonshot launched Kimi K3 in Kimi products and API with 1M context, native multimodality, KDA/AttnRes, and weights promised by July 27. Benchmarks place it near frontier systems, but testers cite slow serving and usability caveats.
OpenAI updates ChatGPT desktop with history sync after Work feedback
OpenAI shipped ChatGPT desktop changes for conversation history, project sidebar access, cross-device Chat and Work history sync, and clearer mode switching. Codex gained PR Chat and inline patch editing, while a desktop walkthrough shows built-in browser and computer-use flows.
Moonshot launches Kimi K3 with 2.8T parameters and 1M context
Moonshot launched Kimi K3 in Kimi products and API with 1M context, native multimodality, KDA/AttnRes, and weights promised by July 27. Benchmarks place it near frontier systems, but testers cite slow serving and usability caveats.
Baseten reports LLM fact writes can retain lift but vanish from answers
Baseten reports that injected facts can retain log-probability lift in weights while disappearing from answers, with prompts recovering 77–80% accuracy. OpenWiki adopted Open Knowledge Format for portable memory files.
OpenAI traces Codex file deletions to $HOME handling bug
OpenAI says GPT-5.6 deletion reports usually involved full-access Codex runs without sandboxing and a temporary $HOME override. Claude Code 2.1.212 also added loop caps and safety fixes.
OpenAI updates ChatGPT desktop with history sync after Work feedback
OpenAI shipped ChatGPT desktop changes for conversation history, project sidebar access, cross-device Chat and Work history sync, and clearer mode switching. Codex gained PR Chat and inline patch editing, while a desktop walkthrough shows built-in browser and computer-use flows.
LocalLLaMA users report near-6x Qwen 3.6 27B speedups with MTP
A LocalLLaMA benchmark on Qwen 3.6 27B and RTX 6000 PRO reports near-6x speedups from MTP, DFlash, and n-gram drafting. Related tests cover remote prefill, NUMA offload, GPU clock tuning, and llama.cpp Gemma 4 support.
AI21 reports 80.8% on SWE-Bench Pro with model-team agent
AI21 says its pipeline routes exploration, extraction, and patching across model tiers and reaches 80.8% on SWE-Bench Pro at $5.99 per task. Other workflows use critic agents, readable harnesses, Fugu/Nemotron routing, and scheduled Gemini agents.
Inkling adds early llama.cpp serving via 1-bit GGUF
Inkling's 1-bit GGUF ran in llama.cpp at 30–40 TPS, and TokenSpeed added day-zero support with a flat KV cache pool. Arena posts put Inkling #10 among open models in frontend code and text, while docs drew scrutiny.
Top storiesthis week
Thinking Machines releases Inkling: 975B open-weight multimodal MoE
Thinking Machines released Inkling with Apache 2.0 weights, 975B parameters, 41B active parameters, text/image/audio support, and up to 1M context. vLLM, SGLang, Modal, Databricks, and Vercel added day-zero support.
Users report Kivine on LMArena may be a Kimi K3 preview
Testers say Kivine identifies with Moonshot/Kimi and produces strong frontend, coding, and spatial demos. Moonshot also teased Kimi K3, but the Arena claims remain unofficial.
OpenAI introduces GPT-Red for prompt-injection red teaming
OpenAI described GPT-Red as an automated red-teaming model for finding prompt-injection vulnerabilities. Posts say it was used in self-play-style training to improve GPT-5.6 robustness.
OpenAI and Work Louder release $230 Codex Micro control deck
OpenAI and Work Louder released kbd-1.0-codex-micro, a compact hardware controller for Codex workflows with mappable buttons, a joystick, status keys, and RGB feedback. Posts from Work Louder and early users describe Codex-focused mappings for multi-agent status, accept/reject actions, six layers, and app auto-switching.
Cursor users report agents switching to costly Claude Opus, Sonnet, Fable, or API calls
Cursor users reported unintended Claude Opus, Sonnet, Fable, or API calls after selecting other settings. Reports included 22.6M-token burns, hard usage stops, and unexpected bills.