1093 musecases built with Meta Muse

Joshua Franco

therealjfranco.itch.io

Peggle/Papa Pear style 2D arcade game. Drop orbs, bounce off pegs, clear levels. Unfinished just-for-fun project built with my AI agent. Free to play in browser or Android APK.

Bill D'Alessandro

@BillDA

Everyone hates taxes. But almost as bad as paying them is finding all the documentation - I dread it every year. My Muse agent just gathered nearly all of it from a single prompt. I'm so happy I might weep 😂 What would have taken me hours, about 90% complete, gathered in a

sequelbox

sequelbox

sequelbox's Tachibana-Agent fine-tune of Muse Glimmer 30B, part of a series also released for Gemma 4 12B and Qwen3.6 27B.

@Satgoy152

@Satgoy152

A project that fine-tunes a DSpark speculator for Muse Glimmer 30B on on-policy coding and agentic traces to raise acceptance length on agentic workloads, evaluated with Terminal-Bench.

@brad-richardson

@brad-richardson

An experimental Rust and Metal inference runtime for Muse models on Apple Silicon, currently supporting Muse Glimmer, using llama.cpp as the correctness and speed baseline.

Alexandr Wang

@alexandr_wang

1/ we’ve built a lot of connectors to make it easier to integrate Muse into your life. Muse runs in a secure VM and we don’t use your data for ads or anything other than to make Muse great for you. you can connect Muse to Gmail, Google calendar, Outlook, Plaid, Opentable, Google

developer.nvidia.com

developer.nvidia.com

NVIDIA's technical blog reports Muse Glimmer serving over 20K tokens/sec on a single Blackwell Ultra GPU and covers RTX 5090, DGX Spark, DGX Station and Jetson deployments.

BeefyDan

@B33fyDan

Progress on the VR Forge: The Agent can now move in its own and grab objects, lift them, carry them and change its colors. I am currently using #astra and #musecode both working in different branches. Soon: Minecraft-like experience building with an Agent. Stay tuned!

AICodeKing tests Meta's Muse Spark 1.2 and the Muse Code terminal agent on the free tier, covering pricing and features before running KingBench 3.

motionlabs.agency

motionlabs.agency

Motion Labs explains Muse Spark 1.2 setup, the cheaper Contributor tier at $0.10/M input tokens, Artificial Analysis scores and using it for content.

Alex Volkov

@altryne

Oh wow, really. I was skeptical but I gave @muse and Instinct the same task, to research tickets for a rosh-ha-shana dinner. @muse is already tightly integrated with link (native integration not just a skill) and it booked me tickets, while Instinct is still searching for the

+1

@caicaitou001

@caicaitou001

A config and write-up that makes muse-spark-1.2-contributor via OpenCode Go work in DeepSeek Harness, fixing empty first-turn output and multi-turn thinking replay errors.

PIMX_ELTEX runs Muse Spark 1.3 in OpenCode on six prompts, from 3D games and an open-world mini GTA to car simulators and web apps, at about 3 minutes per project.

@AstroxNetwork

@AstroxNetwork

An Apache-2.0 template for Meta Muse connectors with per-connection tokens, async job scaffolding and a single registry that generates routes, OpenAPI and llms.txt, deployable to Vercel, Workers or Docker.

TrevorJS

TrevorJS

Refusal-removed Muse Glimmer 30B; the card reports refusals falling from 128/150 (85.3%) to 3/150 (2.0%) on harmful prompts with 0/75 over-refusals.

@accretional

@accretional

Unofficial knowledge base and tools for Meta's Muse Spark, starting with a root-cause write-up of Codex tool-calling failures with Muse Spark 1.1.

Zack Gabbard

@zac70182

@Muse is fantastic! Very impressed. @finkd After importing memory, I asked it to build a personal trainer and it went full systems-engineer on it. It First asked me a series of questions related to my health, gym access, limitations etc. Then it built a program based off of my

@jason-snell

@jason-snell

windows-cua is a native C++ Win32 port of Meta's macOS computer-use agent that drives Muse Spark 1.1 through the Responses API to see and operate a real Windows desktop.

Every model gets the same thing: the Game Boy screen, eleven buttons, a walkability grid, and 1,000 turns to beat Brock. No walkthrough, no pathfinding, no hints. Same prompt for all of them. Scoring is a 10-milestone ladder (left the house, got a starter, reached Route 1, ... beat Brock), so a run that never wins still places by how far it got. 14 runs so far, 4 of them got the badge. Best is GPT-6 Astra at 246 turns. Cheapest win was Gemini 3.8 Flash at $3.70. A free local model (Muse Glimmer 30B) burned all 1,000 turns and got as far as Reached Route 1. Full turn-by-turn logs, costs, and the VODs are all up at https://pokebench.tv.

@emonibnmustafa

@emonibnmustafa

A local translation gateway and profile manager that plugs Meta Muse Spark (or DeepSeek, Groq, Ollama) into OpenAI Codex and ChatGPT Desktop, translating tool calls and SSE streams, with a one-command switch between providers.

cyankiwi

cyankiwi

AWQ INT4 quant of Muse Glimmer 30B calibrated on STEM and agentic data across ten languages, 24.02 GB.

Chris Abraham 🐝

@chrsabraham

I just built the largest collection of helpful prompts (500+) to turn Muse into the best chief of staff, best marketer, best PM you've ever had. @Muse can be so much more than a glorified admin assistant. There are hundreds of ways to use it at your job. Completely free, check

Peter James

@heypeterjames

By chatting normally with the Muse iOS app I was able to have it send a zip of its entire filesystem from the linux root to my Google Drive via the google connector. /home/hatch is the agents home and contains top level files live soul.md, identity.md, memory, tools, and more

I'm not even kidding. It downloaded a fucking emulator and - while I had to provide the rom myself, it's playing Sonic the Hedgehog right now. We're in the endgame, we have to be.

@cellear

@cellear

A read-only visual companion that reads Muse Code session event logs and rebuilds multi-agent sessions (a primary agent and its workers) as a live tree.

@itsgoharr

@itsgoharr

Threadwell Slack is an open-source Muse connector that bridges to Slack's official MCP server via a local OAuth bridge, submitted to the Muse Connector Platform with a full submission pack.

@JacobStephens2

@JacobStephens2

A reproducible evaluation of a stochastic defect where Muse Spark 1.1, under some agent-like request envelopes, wrote to claude-smoke.txt when asked for muse-smoke.txt; a neutral control kept 80/80 filenames intact.

mfgcvguy

@AbdulMathin07

Used Muse to handle a Delta refund end-to-end. Asked it to check the status. It looked up all 3 tickets, opened a live chat with a Delta agent, confirmed the $1,017.60 is going back to my card, and set up daily monitoring until it lands. No hold music involved. @muse

leo

@leo1yu

my @Muse now dresses himself every morning

@mpfaffenberger

@mpfaffenberger

A reproducible single-GPU deployment of Muse Glimmer 30B in BF16 with DFlash speculative decoding on a 96GB RTX PRO 6000 Blackwell, served via vLLM with pinned overlays and smoke tests.

taylorarndt.substack.com

taylorarndt.substack.com

Taylor Arndt tested Muse with VoiceOver and found non-standard text fields and missing heading structure in chats, plus connector gaps such as iCloud email that kept it out of her work.

marktechpost.com

marktechpost.com

MarkTechPost summarizes Meta's numbers: 75.4 on DeepSWE v1.1 (Opus 5 74.0, GPT-5.6 Sol 72.7), 88.8 on Terminal-Bench 2.1, and 98.1 on MRCR v2 at 512K–1M context.

@Israelmusondaayliffe

@Israelmusondaayliffe

Muse workspace skills ported from a Codex/Claude Code plugin marketplace, translating manifests and hooks into procedures Muse can run with its terminal, browser and files.

Hey Guys, I've got 4 x v100's in a Dell C4140 (NVlink) and I have been working on a fork of llama.cpp that is targeted at the v100's. Looking for testers to give it a go and provide feedback. WyvernTKC/llama.cpp-4xV100: Fork of llama.cpp Nvida Volta V100 (tensor parallelism 4 x v100 GPU) model arch type size (GB) pp layer pp tensor change tg layer tg tensor change glm4 9B Q8_0 glm4 dense 9.3 1187.9 2924.3 +146 % 67.6 123.0 +82 % qwen35 27B Q8_K_P qwen35 dense 29.3 640.4 1717.6 +168 % 22.2 52.4 +137 % gemma4 31B Q8_0 gemma4 dense 30.4 679.8 1621.2 ±322 noisy 20.6 46.7 +127 % muse-glimmer 30B F16 muse-glimmer dense 51.9 1048.4 2395.6 +128 % 15.1 40.4 +168 % llama 70B Q8_0 llama dense 69.8 302.5 950.5 +214 % 9.8 28.6 +192 % qwen35moe 35B-A3B Q8_0 qwen35moe MoE 256×8 34.4 1602.6 3143.7 +96 % 93.6 113.7 +21 % qwen3next 80B-A3B Q4_K_M qwen3next MoE 512×10 45.9 889.8 1647.0 +85 % 76.6 86.8 +13 % deepseek4 284B Q2_K deepseek4 MoE 256×6 90.9 188.8 616.1 +226 % 27.3 37.7 +38 % Thanks!

joe feeney

@iamsomewalrus

Very cool experience with Meta Muse. My wife and I were chatting about college tours on the way to get food. I whipped out Muse, listed the schools. Muse did the obvious thing and found the windows for booking etc but, surprisingly, it generated a PDF! Super cool and it was

@syamimhazmi

@syamimhazmi

A vim-modal multi-tab Rust TUI for coding agents where each tab is a session, fronting Muse Spark via muse serve alongside Codex, Antigravity and Grok backends.

@jeperkins4

@jeperkins4

A zero-dependency Ruby client for Meta's Model API focused on the Responses API: reasoning replay across tool loops, previous_response_id state, web_search grounding and typed errors with retries.

Jon ONeill

@HouseHackerJon

Its been 1 day and I had added almost 100 features to 4 apps im building while I slept 300m tokens used Muse QA’s them and pushed to prod as well If you want 1B tokens free sign up here Redeem my code in Settings within 48 hours of joining and we'll both get 1 billion @Muse

mouse.dev

mouse.dev

Mouse asked Muse to archive the files it could see and send them to Google Drive; the agent complied and sent back 6.8 GB of its runtime.

@troioi-vn

@troioi-vn

delegate-to-muse is a Claude Code skill for handing repository work to Muse Code via muse exec, with a handoff template and checks for reviewing the diff and re-running tests rather than trusting the summary.