Entries matching “cars”

ollama

@ollama

Using @AIatMeta's Muse Glimmer all locally to process personal monthly credit card statements. Your data belongs to you! Try different agent tasks using your favorite apps / harnesses with Ollama.

Arun Cherian

@ArunCherian12

Find me a credit card with 0 balance transfer fees, Muse found me a few credit union cards that offer this

boymanrobshit

@boymanrobshit

Getting a ton of utility out of Muse. This may take a minute to catch on because it’s truly a modality shift but here is what I used it for so far. - optimized my credit card rewards. No less than 1000 bucks saved so far. I have five expensive credit cards. I have no idea

Christian Van Der Henst✨

@cvander

Muse just got me tickets navigating a Basque shopping cart. I’m giving meta the lead in the game of agents as of today. 👏

Yair Savlevi

@Yairsavlevi

Landed in Boston with a long layover. My Muse know what cards I own, know where my connecting gates are, made this 3 hour layover a little less painful 😣

Abdul

@AbdulTheBuilder

Big fan of @Muse Took grok bot & ran with it, haven't tried any serious work with it yet but it's working on getting me a refund from enterprise car rentals, booked me a flight for a Christmas trip & found me some pretty sick deals on recliners all within less than an hour of

Instacart

@Instacart

Instacart is coming to @Muse! Soon, you can connect Instacart, say “Taco Tuesday,” and turn the idea into a cart from your favorite store. Then check out and get groceries delivered to your door. Same Instacart grocery experience, new way to get started. tr.ee/6xWsKG

@NeedsChloesure

@NeedsChloesure

Muse Proxy is a Cloudflare Worker gateway that stores CalDAV/CardDAV credentials once and issues scoped API keys, so the Muse agent can reach self-hosted calendars without holding passwords.

@carlosrymer

@carlosrymer

A browser-only agent console for Muse Glimmer 30B that renders the model's ATEM tool-call format client-side, streams from a local server and runs every tool inside the tab across seven capability tracks.

Invest-4-Tomorrow with Kaye

@InvestKaye

Well... that escalated quickly. 😂 Less than 24 hours after posting this, I actually gave Muse the job. “Negotiate my Verizon internet bill.” It logged into my account, reviewed the existing discounts, chatted with Verizon and negotiated: $89.99 → $69.99/month for 12 months

Mike

@MikeAISales

@Muse just helped me book flights for my euro trip. @Ryanair is a nightmare for Americans FYI and muse suggested we try going through PayPal because there probably blocking my American card and then it did the whole PayPal think and booked!

moonShot

@moonShot_6

@meta @Muse has already helped me book a CA DMV appointment for wife yesterday, today it renewed my car registration, scanned my email for open items and create calendar events for others. Crazyy usecases.🔥 always dreamed of having this. @alexandr_wang need more ideas 😁

silv

@mattsilv

i used it to build a little web app to teach my 3 year old the alphabet, with visuals and touch drawing muse.ai/s/letter-cards…

I'm just using OpenWebUI with a simple FastMCP server. Every other model I've tried will simply run a few lines of Python and give me the result. Glimmer seems to overthink like crazy to the point of being useless. On the carwash test it tried to compute emissions using Python. I'm using the recommended sampling parameters, default template, and I've tried both unsloth's Q6_K_XL and Meta's dynamic GGUFs. Any ideas? EDIT: It seems like it's definitely related to the tools available. With them disabled, it's reasonably efficient. I guess it's just overly eager to call every tool it can unlike Qwen or Gemma in my experience.

Disong

@Dis0ng

I'm more of a "casual user", but today I asked Muse to browse for the best deals on an iPhone 17e while I worked on something more pressing. It let me know when it finished, made a widget in our chat showing each promo, and told me which carriers it recommended. Neat!

Boz

@boztank

Very excited about the launch of @Muse today. I’ve been using it internally for months and I am hard pressed to think of any product that I’ve come to rely on more in such a short period of time. I have it linked to my email, calendar, and credit cards. I use it to help me plan

Pete

@petehalvorsen

I should change his name to $cout because he’s saving me money - killed Squarespace, migrated to Shopify - found my Apple TV was already free thru my Chase Sapphire Reserve, activated that one - hostgator has the reaper coming for it. Bud Fox my finance guy making moves. Iykyk

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!

A static Space showing a Hermes-agentic fine-tune of Muse Glimmer 30B proposing one or two tool calls or answering directly; the model card reports 16/20 on a simulated loop gate.

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.

Qubicle

@cubeqube

Got my Muse setup and so far - it’s managed my 401k for me, rebalanced it, changed allocations to save me fees on some similar funds, put a dashboard together of all my financial accounts and transactions and set up budgets and recurring subscriptions as well as tracking my

Quick update on the RTX 5060 Ti local LLM repo. It has changed quite a bit since my previous posts. The project started as a collection of practical notes and benchmark results. That was useful, but as the dataset grew it became harder to answer the question most people actually had: What configuration should I run? I have rebuilt the repo around tested, copyable presets rather than treating every successful benchmark request as a front-page result. What changed? The project now separates three things: • Presets: exact configurations intended for people to copy and run. • Evidence bundles: reviewed proof of context fit, retrieval, sustained generation and performance. • Raw receipts: retries, failed experiments and diagnostic runs that are kept separate as engineering material without automatically becoming recommendations. The website now leads with the published preset catalogue. The larger results explorer is still there for comparisons and historical data, but it is no longer the first thing visitors have to decipher. There are currently seven published presets across the 1× and 2× RTX 5060 Ti lanes: 1× RTX 5060 Ti 16GB • Qwen3.8 27B IQ3_XXS at 64K with q8 K

HealthEx launched a connector that lets Americans bring their medical records to Muse, which can then write visit summaries, draft questions for a doctor and set refill reminders.

An independent catalog of 11 Telep IO connectors for Muse, covering package tracking, video summaries, postal mail, faxes, e-signatures, shipping labels, domain registration and robot-handwritten letters.

Meta's official Muse-Glimmer-30B repo: ~29.6B dense model with a 1.8B vision encoder, 131K context, Apache 2.0, with vLLM and SGLang serve commands.

Stripe integrated Link's agent wallet into Muse: at 1M+ Link-accepting businesses Muse checks out with the user's saved method, elsewhere Link issues a single-use virtual card scoped to the purchase.

Frank Jerome ⚡

@frankilus

.@Muse just saved me $603 a year on my car insurance after analyzing my policy. This assistant is really good at getting things done.

shashank

@_shanxS

I managed my car purchase: research, haggling with dealership, finding right insurance quote, booking pre-purchase inspection etc

Joshua Liu

@joshuapliu

Meta’s Muse AI Agent already saved me $113/month on car insurance - no joke… BUT I fear agents will expand the Digital Divide for patients. Older patients, less tech savvy patients, lower SES patients already face barriers accessing care. When the most tech savvy consumers now

Full Value Dan

@fullvaluedan

I'm really surprised with what you can do with @Muse Most agents struggle with making decent images. For marketers, Muse automate can: - Create detailed carousels with custom images - Create skills and verification steps to learn your style - Post to social media - DM people

@Jemno23

@Jemno23

A Vercel-hosted MCP connector for Muse that pushes cards to a 600x600 display web app, built around a serverless socket-lifetime spike with session state kept in memory or Redis.

Minh Phan

@MinhInvest

use muse from $META today and the agent helped me getting 4 different car insurance quotes while i go for groceries shopping. i ended up switching my car insurance from tesla to CSAA from $500 to $335 per month the same day

Just had Muse confirm what I can use a Verizon gift card for. I didn’t realize I could apply it to my account balance. The agent then pulled up Verizon.com and processed the payment for me. I was about to buy a shitty phone case and sit on the balance… just saved me $100.

This is a small HTTP API you run on a machine you control. Muse SSHes in and calls http://127.0.0.1:8792. It can search Amazon and add to a personal cart or a Prime Business cart. It does not check out or pay. https://github.com/aasper03/muse-amazon-bypass You need Python 3.11 and curl. Sign in to Amazon in a normal browser and export cookies for each account. Amazon blocks automated login windows. The README covers Muse-over-SSH setup, and a later section shows how to call the same API yourself.

Erik Lammerding

@Lammerding

@muse found me a car ferry from Croatia to Italy. Saved me $2200 in drop-off fees. Booked/paid the ferry, bonus - extracted information like license plate, car model, etc, from the rental agreement, confirmed I was insured for the voyage. 🙌@alexandr_wang @natfriedman

Mark Zuckerberg

@finkd

Workflows can break out a task across multiple focused agents, carry the intermediate work between stages, and return one result at the end. Good for engineering work that's too complex for a single agent.

Jerrod Lew

@jerrod_lew

Meta Muse app has been launched. I asked it to do full online research of myself and it created a full portfolio! Included a career timeline, social media platform stats and all of my speaking arrangements. This is a great way for a new agent to learn about you.

james

@ads4apps

Having @Muse find the best offer to sell my Model X. Then it’s finding a lightly used new version of the same car

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.

@Codys12

@Codys12

A correctness-first bring-up of Muse Glimmer 30B on a single Tenstorrent p150 card, with paged KV cache, native DFlash speculative decoding and an OpenAI-compatible server.

A Muse Glimmer fine-tune that writes, tunes and translates detection rules (Sigma to KQL and SPL, YARA, Wazuh) and runs 4-bit on a single 24 GB card.

Idobn

@idobn

Had @Muse go through 6 months of my email receipts. It found a $100/mo Claude Max sub on my corp card I'd forgotten about, while we already have an enterprise license. $1,200/year saved. @finkd clearly articulates the biz model here - muse is supposed to save / make you money,

@terramike

@terramike

A Muse skill that helps people buy crypto (XRP by default) with a card through Changelly, giving CoinGecko spot estimates, prefilled buy links and plain fee disclosure without any API keys.

Abhiraj Butala

@abhirajbutala

Asked my @muse to find the items from a saved IG reel. It "watched" the reel, matched everything across Amazon & Etsy, and loaded my carts with a few solid options per item. Spared me the research grind entirely.

@harris-ryder

@harris-ryder

Dowser is a read-only MCP connector for Muse that surfaces open class-action settlements, product recalls with refunds, unused card credits, expiring points and birthday freebies, each with an official link.

@amanbthakkar

@amanbthakkar

An open protocol plus Muse skill and Cloudflare Workers reference server that lets each family member's Muse join shared boards (groceries, trips, household lists) with per-board identities and an approval card before anything is ordered.

Hey. Just tried it on my old ass gpus 😄 Surprisingly Tensor Split is working on 2 gpus almost doubling PP (wonder how it will work with 4 gpus) Q6 — 1 GPU llama-server \ --model <MODEL_DIR>/Muse-Glimmer-30B-GGUF/Muse-Glimmer-30B-UD-Q6_K_XL.gguf \ --mmproj <MODEL_DIR>/Muse-Glimmer-30B-GGUF/mmproj-kquant.gguf \ --spec-draft-model <MODEL_DIR>/Muse-Glimmer-30B-GGUF/dflash-kquant.gguf \ --spec-type draft-dflash \ --spec-draft-ngl 999 \ --spec-draft-n-max 3 \ --spec-draft-type-k f16 \ --spec-draft-type-v f16 \ --ctx-size 65536 \ --override-kv muse-glimmer.context_length=int:65536,dflash.context_length=int:65536 \ --n-gpu-layers 999 \ --device ROCm0 \ --device-draft ROCm0 \ --split-mode layer \ --flash-attn on \ --fit off \ --parallel 1 \ --kv-unified \ --batch-size 2048 \ --ubatch-size 512 \ --threads 32 \ --threads-batch 32 \ --cache-type-k f16 \ --cache-type-v f16 \ --image-min-tokens 1024 \ --image-max-tokens 4096 \ --reasoning-preserve \ --temp 0.7 \ --top-p 0.95 \ --top-k 64 \ --min-p 0.0 \ --jinja Q8 — 2 GPUs with tensor split bash llama-server \ --model <MODEL_DIR>/Muse-Glimmer-30B-GGUF/Muse-Glimmer-30B-UD-Q8_K_XL.gguf \ --mmproj <MODEL_DIR>/Muse-Glimmer-30B-GGUF/mmproj-kquan

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

Nick Prince🛡

@Nickprince

muse saved me over $1k yesterday was buying out my leased car and the dealership offered to handle all the hassle around the loan and dmv what they didn’t say was they were hiding an enormous fee for doing so something looked off and muse explained exactly what was going on in

Instacart is bringing personalized grocery shopping to Muse through its connector program, drawing on real-time inventory from about 100,000 stores.

Eleven matched on/off pairs across Gemma 4 and Qwen3.6, holding model, quant, card, corpus and concurrency fixed inside each pair. Speed: 1.65x to 2.54x, every pair. Accuracy: nothing the paired intervals could separate from ordinary run-to-run movement. Muse Glimmer is the one that lost. Meta's matching DFlash drafter made the same 7900 XTX 9% slower, keeping 24.55% of drafted tokens against roughly four in five for the Gemma and Qwen heads. Acceptance fell across the run instead of warming up. Meta's model card reports 3.1x on an RTX 5090, and there are open llama.cpp issues for DFlash on AMD and under Vulkan, so I read it as the backend rather than the model. Acceptance turned out to be a poor predictor of speed. It moved under four points across five models while the multiple nearly doubled. What tracks the multiple is how bandwidth-bound the target is: a heavier quant gains more, and the two mixture-of-experts pairs gained least. Worth knowing before you benchmark anything: -md mtp-head.gguf silently disables speculation. Use -hf REPO:QUANT -hfd REPO, then read speculative from /slots and confirm it is true. Per-pair table, intervals, acceptance counters and the raw predic

Same old prompt, just appended a TIP in the end: "Write a single HTML file with a full-page canvas and no libraries. Simulate a realistic side-view of a moving car as the main subject. Keep the car visible in the foreground while the background landscape scrolls continuously to create the feeling that the car is driving forward. Use layered scenery for depth: nearby ground, roadside elements, trees, poles, and distant hills or mountains should move at different speeds for a natural parallax effect. Animate the wheels spinning realistically and add subtle body motion so the car feels connected to the road. Let the environment pass smoothly behind it, with repeating but varied scenery that makes the movement feel believable. Use cinematic lighting and a cohesive sky, such as sunset, dusk, or daylight, to enhance atmosphere. The overall motion should feel calm, immersive, and realistic, with a seamless looping animation. TIPS: You don't have vision abilities so don't try it yourself. If you feel in need of vision ability, you can access http://xxx:8080/v1, model id: Muse-Glimmer for help, it will see the picture, and describe it for you." Then the PI agent started spinning, round a