@bnjmn_marie
Independent AI researcher (LLM, NLP). My blog, The Kaitchup - AI on a Budget: https://kaitchup.substack.com/
AI
In a September 16 update, the team behind MiMo-V2.6 said it was scaling compute, agent environments and grading, with plans to open-source details in stages.
AI
In one user's long-context coding tests on an RTX Pro 6000, NVIDIA's version with MTP-4 was roughly 2.5–3× slower than Unsloth, despite being about 2× faster than running without MTP.
AI
A user calls V4.1 “very brittle,” claiming it can see “massive gains” when AGENTS.md instructions are optimized for a particular scenario.
AI
The Next Web says OpenAI published the higher score, while the benchmark’s own software returned the lower one. The model and test were the same; the surrounding software setup differed.
AI
Independent researcher notes persistence of older evaluation method for advanced AI models.
AI
Pseudonymous AI commentator @scaling01 asserts the claim about the model in a post.
AI
Artificial Analysis calls out its edge on cost per intelligence task.
