prinz
@deredleritt3r · Oct 2
Claims the model decoded two never-before-read letters from the secretary of the Duke of Mayenne to the Pope's nephew.
61 entries across 11 products
Research & posts
Benchmarks, write-ups and discussions. 61 entries for research & posts from 55 creators on X, GitHub and Web, published Aug 11 to Oct 7.
Entries
@danshipper
@danshipper · Sep 29
Dan Shipper relays OpenAI's Space, a Drive-like area in ChatGPT where people and dots share files, plus native docs, sheets and slides apps.
Alexandr Wang
@alexandr_wang · Oct 2
Wang shows Meta own Home Link bridge device that connects Muse to smart-home gear, produced in a limited batch of 5,000.
Peter Yang
@petergyang · Sep 26
Teaser for a deep dive with the engineering and design leads on the 14 bots they use, including a design bot, with a preview image.
Arena.ai
@arena · Sep 30
Arena places Argon (High) at #8 in its Agent Arena with a positive net-improvement score and about $0.62 per task, a roughly 5-point gain over Gemini 3.8.
eric zakariasson
@ericzakariasson · Sep 30
Cursor engineer points to the engineering team's bots now installable from the marketplace, with a screenshot of the listing.
ハカセ アイ
@ai_hakase_ · Sep 26
Japanese explainer of the lazy dequantisation and expert cache design, with about 65 tok/s.
Salio
@Mr_Salio · Oct 1
Measurements from 24 hours of testing showing Sol 6.1 much slower than GPT-6 Sol and GPT-5.6 Sol at xhigh.
hku-sail
@hku-sail · Sep 7
HKU study that rebuilds robot scenes from three RGB views and replays the actions in Blender, with editable replay files.
ollama
@ollama · Aug 14
Demo clip of ollama launch dsh running the harness fully locally with Ollama web search preinstalled and a trajectory view.
@tenobrus
@tenobrus · Sep 30
Funny screenshot of a dot querying OpenAI's help-site support chatbot to learn about its own capabilities.
Captain Sude
@captain_sude · Sep 16
Post claiming Astra produced a full proof of a Liouville variant of the Goldbach conjecture that was then verified in Lean.
Patrick Wardle
@pwardle · Sep 21
Patrick Wardle's security PoC showing how an unprivileged local process could redirect Muse's dictation traffic.
Dan McAteer
@daniel_mac8 · Oct 1
A free skill that helps Claude write and package Claude Code mods, released on launch day by Dan McAteer.
fulvian
@fulvian · Sep 5
Experiment that edits the model's n-gram embedding table to inject removable facts without fine-tuning.
Daniel Nguyen
@daniel_nguyenx · Sep 30
Scored comparison of coding output where Sol 6.1 lands near Grok 4.7 behind Astra and Opus 5.5.
David Hendrickson
@TeksEdge · Oct 1
Experiment pairing GPT-6.1 Sol as planner with a local Qwen3.8-27B coder on an RTX 3090, with timings and costs.
Jesse Vincent
@obra · Sep 29
Jesse Vincent tracks every skill file in the Hatch container with a commit per day so changes to Muse's built-in skills are visible.
jacobwell
@jacobwell · Sep 25
Jacob Well published the documented SKILL.md for each built-in Muse skill, from Gmail to Hue to Spotify, as a reference index.
Geek
@geekbb · Aug 15
Post collecting several community desktop front ends for DeepSeek Harness, with an image of the first one.
Louis-François Bouchard
@Whats_AI · Sep 29
Louis-Francois Bouchard's internal benchmark for GPT-6.1 Sol across effort levels with Elo, cost and time per script.
