Models & Labs
GPT-6 Astra Edits Qwen Weights Directly
Lev Selectorhigh confidence
Why it matters
- →Demonstrates direct weight manipulation by LLMs
- →Signals shift toward autonomous model optimization
- →Raises safety concerns regarding open-weight integrity
In a demonstration of advanced model interaction, OpenAI's GPT-6 Astra was shown directly editing the weights of Qwen models. This capability suggests a shift toward autonomous model refinement, where larger or more capable models can optimize smaller ones without human intervention. The event highlights the emerging trend of 'models editing models,' raising questions about safety, control, and the future of open-weight ecosystems.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
How we got here
- 1OpenAI Unveils GPT-5.6 with Sol Model
Fireship · July 10, 2026 · Related
- 2OpenAI Unveils GPT-6 Astra Model
OpenAI · September 3, 2026 · Same story
- 3OpenAI Unveils GPT-6 Astra, Claims AGI Era
The Verge AI · September 3, 2026 · Related
- 4OpenAI Launches GPT-6 Astra, Claims AGI Milestone
WIRED AI · September 3, 2026 · Related
- 5OpenAI unveils GPT-6 Astra, claims AGI milestone
The Rundown AI · September 4, 2026 · Related
- 6OpenAI Releases GPT-6 Astra with New Safety Paper
AI Explained · September 4, 2026 · Related
- 7OpenAI Unveils GPT-6 Astra
Matt Wolfe · September 4, 2026 · Related
- 8OpenAI Unveils GPT-6 Astra for Business Use
OpenAI · September 9, 2026 · Related
Follow this story
Open the full story →MIT shows AI rewriting model weights without training
2 developments
- Nov 3 · The Rundown AI
MIT shows AI rewriting model weights without training - Nov 6 · Lev Selector
GPT-6 Astra Edits Qwen Weights Directly (This article)↳ OpenAI's GPT-6 directly edits Qwen weights, advancing AI self-improvement beyond MIT's training-free method.
More from Lev Selector
© Lev SelectorVideo & Creative AIimage
OpenAI Images 2.6 Released
OpenAI has updated its image generation model to version 2.6, bringing improvements in quality and consistency.
Lev Selector
© Lev SelectorOpen Sourceimage
FLUX 3 Image Klein Open Weights Released
Black Forest Labs released FLUX 3 Image Klein, an open-weight image generation model offering high-quality results for developers.
Lev Selector
© Lev SelectorGeneral AIagents
Siri AI Opens to Developers
Apple has opened its Siri AI capabilities to third-party developers, expanding the ecosystem of voice-activated integrations.
Lev Selector
More in Models & Labs
Models & Labsother
llama.cpp b12177 adds CUDA 13 and ROCm 10
This release quietly closes the hardware gap for local inference by adding default support for CUDA 13.5 and ROCm 10.0 alongside existing CUDA 12 builds. AMD GPU users finally get first-class parity with NVIDIA, while Snapdragon mobile chips gain dedicated NPU acceleration paths. Apple Silicon KleidiAI is explicitly disabled here, a notable regression for Mac performance that suggests ongoing tuning issues. The inclusion of OpenVINO and SYCL backends further cements llama.cpp as the universal runtime for inference across nearly every major silicon architecture.
llama.cpp Releases
Models & Labsmodels
llama.cpp adds Kimi K3.1-Pro support
Kimi’s new K3.1-Pro model joins the llama.cpp ecosystem, bringing a distinct architectural twist to local inference. By replacing full-attention MLA layers with DSA (Dynamic Sparse Attention) and linear-attention KDA layers, this release tests how well sparse caching handles hybrid attention patterns. The implementation keeps indexer precision high to avoid quality loss during top-k selection, ensuring the model runs accurately on consumer hardware. This expands the library’s reach into newer Chinese open-weight models that rely on complex memory management.
llama.cpp Releases
Models & Labsother
llama.cpp b12182 release with CUDA 13 and ROCm 10
This release quietly cements llama.cpp as the universal inference runtime by adding default builds for CUDA 13.5 and ROCm 10.0 across Linux and Windows. The inclusion of ROCm 10 means AMD GPU users no longer need custom compilation to access modern acceleration, effectively closing the gap with NVIDIA's ecosystem support. While KleidiAI on Apple Silicon is disabled in this build, the broader platform coverage ensures that local inference remains accessible on virtually any hardware configuration without manual setup.
llama.cpp Releases