NVIDIA quietly shipped a 4-billion-parameter model called Dory1.0-4B on Hugging Face yesterday — and the unusual part is that one model can run its middle block one, two, or four times per token…
AspireAI community · huggingface.co
NVIDIA quietly shipped a 4-billion-parameter model called Dory1.0-4B on Hugging Face yesterday — and the unusual part is that one model can run its middle block one, two, or four times per token at inference.
Dory is what's called a "looped" or "latent-recurrent" model: the same shared weights are repeated. Bumping the loop count trades more compute per answer for more accuracy, without changing the checkpoint on disk. NVIDIA's vendor numbers put SWE-bench Verified at 50.3 / 59.5 / 61.8 across 1, 2, and 4 loops, within a point of Qwen3.5-9B (61.0) at the top end.
The OpenMDW 1.1 license permits commercial use, and four quantizations shipped day one: BF16, MXFP8, NVFP4, GGUF. The honest limit: no NVIDIA dev-blog launch post surfaced with the weights, so the benchmarks are vendor-only and no third party has independently reproduced them yet.
Read the story and explore more from the AspireAI community.
Two critical PHP remote code execution flaws — one found by GPT-5.6 Sol — were patched on October 21, 2026, and OpenAI published a disclosure framework alongside the fix. Fixed versions: PHP 8.5.12, 8.4.27, 8.3.36 and 8.2.35. Both allow unauthenticated remote code execution, and PHP's advisory says risk is highest for mod_php, where a demonstrated exploit triggers before any application code runs. The CVEs are distinct failure modes. CVE-2026-109618 (OOM corruption): out-of-memory handling can interrupt PHP while internal request data is only partially initialized, and request cleanup may then treat the invalid data as valid. CVE-2026-109619 (stale global state): request-owned internal state can survive after its backing memory is freed, and a later request served by the same worker may use that stale state. OpenAI states GPT-5.6 Sol identified the first flaw through its Patch the Planet initiative, and that Wiz independently discovered and reported both vulnerabilities. PHP's maintainers led the fixes and release, with Wiz, Cloudflare and OpenAI contributing research and coordination. Wiz's write-up adds that it used GPT-5.6 Sol and Gemini 4 Argon throughout, with Gemini 4 Argon contributing to finding the vulnerability and validating the patch. OpenAI describes the gap between discovery and public exploitation as a "Defender's Window," proposes a five-stage response framework, and committed $1 billion in subsidised Daybreak model access for organizations with limited security resources. Honest caveats: these are OpenAI's and Wiz's own characterizations, and OpenAI also notes the same capabilities help attackers build offensive tooling faster. PHP 8.1 and older no longer receive updates from the PHP project.