Writing
All postsBenchmark results for AEON-7/Qwen3.6-35B-A3B-heretic-NVFP4 + DFlash (NVIDIA DGX Spark, GB10)
The previous post rejected MTP-1/2 and projected ~2× from NVFP4. We measured +152% aggregate at c=10 and discovered that published benchmark numbers are only honest when paired with the workload that produced them.
Real Agency Is a Loop, Not a Prompt
An agent that can't come back, is an agent with no agency, a function that has reached its end.
Benchmark results for Qwen/Qwen3.6-35B-A3B-FP8 (NVIDIA DGX Spark, GB10) serving via vLLM
Benchmark results for Qwen/Qwen3.6-35B-A3B-FP8 (NVIDIA DGX Spark, GB10) serving via vLLM.
Open source
GitHubruimachado-orbit/lecta
feat(slides): add MDX support for JSX-powered slides
Runtime MDX compilation via @mdx js/mdx with two layer caching (processor + output). .mdx slides render JSX with inline styles alongside stan...
FasterXML/jackson-databind
Add `JacksonCollectors` with `toArrayNode()` implementation (#4709)
Commit 567e089
apache/commons-configuration
feat/configuration-contains-value (#412)
feat: contains value initial implementation test: tests adjusted chore: remove temp class chore: lint fix fix: MapConfiguration contain...
opensearch-project/opensearch-py-ml
Adds interactive and output path options to model register (#591)
Adds interactive and output path options to model register Enables interactive mode and custom output path selection when registering a model via...
Teams
Maio Labs
Building with Maio Labs on AI product workflows, execution, and systems that need to hold up under real usage.
Eidosmedia
Part of R&D, innovation, and AI, pushing Eidosmedia toward bleeding-edge technology and building AI solutions across the company's wider product horizontals.