GitHub

@ryangu00

@ryangu00

A measured deployment report of Muse Glimmer 30B NVFP4 with DFlash on a single Dell Pro Max GB10, where it posted top vision and SRE-ops scores but failed five deployment gates against DeepSeek V4 Flash.

@Srimi1

@Srimi1

An experimental macOS gateway that keeps the stock Muse Code CLI in charge of the UI, sessions and tools but swaps the model connection for OpenAI or Z.ai models via a loopback gateway.

@nuocmam

@nuocmam

A same-prompt bakeoff site that starts with a playable penguin Tetris generated by Muse Spark 1.3 in OpenCode, with a fixed prompt for comparing other models.

@mgaruccio

@mgaruccio

A vLLM-XPU and DFlash recipe for Muse Glimmer 30B on a single Intel Arc Pro B70, reporting 278 aggregate tok/s across eight clients and an 840.8 tok/s burst peak at concurrency 96.

@dominicletz

@dominicletz

A self-contained visual report comparing CursorBench 4.0 score against cost per task for Opus 5.5, Fable 5.1, Grok 4.7 and Muse Spark 1.3, where Muse Spark 1.3 Max scores 41.6% at $2.64 per task. GPT-6 points are clearly marked as estimates.

@latekaapi

@latekaapi

A native macOS window for Muse Code that drives muse serve over MSP and renders approvals, questions, plans, todos and tool output as interface elements instead of terminal text.

@Solar2004

@Solar2004

A config (in Spanish) plus two helper proxies for using Muse Spark 1.2 Contributor from Claude Code and Hermes through routatic/proxy and an OpenCode Go subscription.

@mapleroyal

@mapleroyal

A local Muse Glimmer 30B vision-and-reasoning chat app for high-memory Apple Silicon Macs, running inference through ExecuTorch, MLX/Metal and DFlash with nothing persisted to disk.

@murpheycandler

@murpheycandler

A small Inspect evaluation on Muse Glimmer that tests whether incentive framing changes what an agent reports to its principal when the evidence is held constant; the author reports a null result.

@fortun8te

@fortun8te

ocodex fans out Muse Spark 1.2 Contributor workers on decomposable coding work while one paid supervisor audits every claim, with crash checkpoints, heartbeats and a machine-readable ledger.

@raksix

@raksix

A documented setup that runs Muse Spark 1.3 Contributor inside Claude Desktop on macOS through a small Node proxy that translates Anthropic requests to OpenCode Go's OpenAI-style API.

@ezcat207

@ezcat207

A curated, source-linked catalog of 156 real things people have done with the Muse personal agent, grouped into 16 categories with every entry linked to its original X post.

@r0b0tlab

@r0b0tlab

Reproducible native NVFP4 serving of Muse Glimmer 30B on one DGX Spark (GB10) with unmerged vLLM support: about 10.3 tok/s single-stream versus 4.2 for BF16, 52.5 tok/s at c16, and 131K context checked with needle-in-a-haystack tests.

Anonymous

github.com

A working mobile-friendly web app built by the Meta Muse personal agent from a real courier's detailed requirements. It compares DoorDash/Uber Eats delivery offers after total miles (including empty return), waiting time, all-in vehicle costs and tolls. The surprising demo: a $9 offer can yield $19.08 net per hour while a $20 offer only yields $16.09 under stated assumptions. Muse created the standalone offline HTML and logic tests (49/49 self-reported); I downloaded it, independently verified actual browser inputs/outputs, published the source and proof screenshot to GitHub, and deployed a no-signup live demo. Live app: https://boyhoangxattpt2-blip.github.io/gig-driver-profit-checker/ . GitHub README includes exact math, source code, build story, and reproducible prompt.

@Aeroknight786

@Aeroknight786

A local browser GUI for an existing Muse Code install, offering a persistent view of projects, tasks, conversations, permissions, activity and working-tree changes over Muse's session protocol.

@meta-models

@meta-models

Meta's official recipes for building agents and coding tools on the Model API with Muse Spark 1.3, covering API basics, multi-agent patterns and end-to-end use cases via OpenAI- and Anthropic-compatible SDKs.

@lobanov

@lobanov

An effort to make Muse Glimmer 30B actually use a 512k-token context (4x native) as a ~17GB GGUF in 32GB VRAM, trained on DGX Spark and evaluated with RULER-style retrieval tests.

@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.

@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.

@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.

@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.

@luckeyfaraday

@luckeyfaraday

An unofficial loopback shim that lets Meta Muse Code use OpenAI Responses APIs, OpenRouter, Codex OAuth and the official Claude Code CLI, switchable from Muse's /model menu.

@bex-co

@bex-co

muse-code-acp is an unofficial Node adapter that exposes Meta's Muse Code through the Agent Client Protocol, so it can be driven from Zed or VS Code with streaming, tool calls and MCP servers.

@everyai-com

@everyai-com

A set of small fixes that let Muse Spark run in pi, Agent Orchestrator, opencode and other OpenAI-compatible tools, handling its non-standard SSE event and missing model-catalog entries.

@cobusgreyling

@cobusgreyling

Cobus Greyling's companion repo for Muse Glimmer 30B pairs a long-form intro with an offline-first interactive lab for exploring agent loops, benchmarks and memory envelopes before downloading the weights.

@karimkfoure

@karimkfoure

An OpenCode plugin that ports Muse Code's subagent coordination core: durable child workers, a ledger mailbox and admission control, so subagents survive crashes and respect capacity.

@srikantmehra57

@srikantmehra57

An independent desktop command center for the Muse Code CLI that brings workspaces, sessions, approvals, reasoning controls, changed files and agent activity into one interface.

@z0zero

@z0zero

A looping browser 3D rocket launch made with Muse Spark 1.3 in Vite, React and plain Three.js. Ignition, liftoff and ascent run about 22 seconds, with all geometry and smoke procedurally generated and no external assets.

mastodon: serrebiradio.com/@serrebi

github.com

It is a frontend and a chat client for AI providers, supporting 7 backends including muse code

@airawatraj

@airawatraj

Inference tuning notes for serving Muse Glimmer 30B NVFP4 with DFlash on a single NVIDIA DGX Spark as a consistent agent backend; the repo reports 27.5 tok/s average and 90/100 on its tool eval with 128K context.

@homerquan

@homerquan

A start/stop/status launcher that serves the NVFP4 Muse Glimmer 30B checkpoint on NVIDIA DGX Spark with vLLM, Glimmer's reasoning and tool parsers, and its DFlash speculative decoder.

@ALeksandr13254

@ALeksandr13254

A fork of NemoAgent that swaps the LLM layer to OpenCode Go models, defaulting every role (dialogue agent, executor, router) to Muse Spark 1.3 Contributor with a per-role reasoning level.

@o1ie12

@o1ie12

A single-file marketing site (hash routing, GSAP transitions, Lenis scroll, Three.js) built by Muse Spark in Meta AI from 5–6 prompts, with no manual code edits.

@nano-muse

@nano-muse

nanoMuse is an open-source personal AI agent inspired by Meta Muse that runs on Android with a shell, browser, MCP, skills and scheduled tasks, and can be pointed at DeepSeek, OpenAI or Ollama models.

@vcruz305

@vcruz305

Tested SGLang and vLLM launch recipes for an NVFP4 export of a Hermes-agentic fine-tune of Muse-Glimmer-30B on NVIDIA DGX Spark (GB10).

@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.

@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.

@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.

@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.

@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.