| ▲ | Our Smallest, Most Efficient Model: Muse Ember(research.meta.ai) |
| 157 points by dvaplima 3 hours ago | 64 comments |
| |
|
| ▲ | simonw 3 hours ago | parent | next [-] |
| Six effort levels (minimal, low, medium, high, xhigh, max), so here are six pelicans (actually five, "high" didn't return a complete SVG): https://tools.simonwillison.net/markdown-svg-renderer?url=ht... There's usually a reasonably strong correlation between thinking level and visual quality of the pelican. This is the first time I've seen that correlation appear to break - more reasoning tokens doesn't seem to produce an obviously better SVG. |
| |
| ▲ | imimayj1337 2 hours ago | parent | next [-] | | It does seem pretty insane to me that only on max setting (and after waiting 27 seconds) did you actually manage to receive an SVG where it appears that the pelican's legs are straddling the bike correctly. Edited to make clear what it looks like to me and what I was looking for. | | |
| ▲ | swatcoder 2 hours ago | parent | next [-] | | They're not. It's an illusion from the composition. If you look at the code for each of these (and probably most across models), you can see that the drawing elements are organized by concept. Presumably, this helps accuracy by keeping attention on one concept at a time, but you end up with a stack of conceptualized objects and lose the ability to interleave things. No doubt these could all do better with a more thorough and specific prompt that coached them away from that technique, but otherwise it's a very subtle detail that a "thinking" process needs to stumble upon independently. | | |
| ▲ | flir 2 hours ago | parent [-] | | You could use a second prompt to z-order the objects (I tend to break down tasks and pipeline them, rather than trying to stuff everything in a one-shot prompt. Seems to work better for me, could be a holdover from the early days). |
| |
| ▲ | latexr 2 hours ago | parent | prev [-] | | They’re not. Note how both pedals are on the same side of the frame. That’s wrong! |
| |
| ▲ | ezst an hour ago | parent | prev | next [-] | | Wouldn't that be a sign that those models were trained to optimise that test case specifically, as in, by exhibiting equal or higher performance while using less power? | |
| ▲ | meric_ 2 hours ago | parent | prev | next [-] | | Wow, minimal seems to be the best here interestingly. Only 10 reasoning tokens so basically none? | |
| ▲ | eterm 2 hours ago | parent | prev | next [-] | | Did minimal just hand it off to something else? It seems completely out of line with the rest. | |
| ▲ | permalac 2 hours ago | parent | prev | next [-] | | Is there a problem if I like the minimal more than the others? | | | |
| ▲ | AlecSchueler an hour ago | parent | prev | next [-] | | Any idea why there's so much convergence on the fish in a basket idea? | | |
| ▲ | swatcoder an hour ago | parent [-] | | Same reason they're all facing right with the same composition, they all have three motion lines, they all have a yellow sun and 2-3 blobby clouds, they're all on two-lane pavement with a white dashed lane separator. And the same reason Claude has all its notorious Claudisms in prose. The shorter and vaguer a prompt you give to an LLM, the more it fills in further details using associations they've developed through training. These are ley lines that the models gravitate towards. Diffusion models too. It's why lazy AI use for creative work always comes across as lazy in the wild and always will. When you use AI for "creative" purposes yourself, it feels like magic to see a pelican riding a bicycle at all, and you're impressed by all those cool details it added. But then 10, or 100, or 10M people throw more or less the same trivial prompts at it and we start to recognize all those "ley line" associations fill in the gaps the same way again and again and again. | | |
| ▲ | viraptor 4 minutes ago | parent | next [-] | | To be fair, people have a genuine preference for images with things facing to the right. And lots of drawings will do 3 motion lines. You'd likely draw a dashed rather than solid separator too, or it would look like a random line. Lots of the AI habits are there because of choices we make. | |
| ▲ | mvkel an hour ago | parent | prev [-] | | This is interesting. There's a balance between: - simple prompt for complex output (bad) - complex prompt for simple output (bad) - simple prompt for simple output (good) - complex prompt for complex output (.... is this good or bad?) |
|
| |
| ▲ | mgdev an hour ago | parent | prev | next [-] | | Fidelity. | |
| ▲ | phyrex 2 hours ago | parent | prev | next [-] | | I like that from xhigh on up its wearing a helmet. | |
| ▲ | sjacob 2 hours ago | parent | prev | next [-] | | Incredible. How do you get these out so fast? | |
| ▲ | jolt42 2 hours ago | parent | prev | next [-] | | Pelican is the new teapot | |
| ▲ | djmips 2 hours ago | parent | prev | next [-] | | birds at minimal and return at max | |
| ▲ | small_model 2 hours ago | parent | prev [-] | | Not sure this is a good way to benchmark LLM's, no offence, but a single SVG is hardly a measure of a models capability. |
|
|
| ▲ | Retro_Dev 19 minutes ago | parent | prev | next [-] |
| When they said the model was the "smallest" of their models yet, I was really hoping it would be open weights. I can't find any weights, how sad :( |
|
| ▲ | kbrannigan 2 hours ago | parent | prev | next [-] |
| This is starting to feel just like the early PC days. A bunch of Computer models coming out, all shapes and color, but eventually it settled around mouse keyboard, monitor most people, use the office suite: Word, Spreadsheet , presentations, database. I bet we're going to see the exact same thing with LLMs at specialized models instead of the everything models. One for lawyers, coders, Accountants, Designers, etc… |
| |
| ▲ | grim_io an hour ago | parent [-] | | I don't think so. Cross discipline skills are more valuable in the real world, outside of hyper specialization seen in the California tech scene. Even there, it's coming to an end. |
|
|
| ▲ | minimaxir an hour ago | parent | prev | next [-] |
| > It costs $0.10 per million input tokens, $0.004 per million cached input tokens and $0.50 per million output tokens, Notable cache cost of 4% input, compared to Luna/Haiku's 10% of input. |
|
| ▲ | open-paren 2 hours ago | parent | prev | next [-] |
| I believe this is what the “crouton alpha” stealth model was https://openrouter.ai/stealth/crouton-alpha |
| |
| ▲ | bityard 2 hours ago | parent [-] | | Based on? | | |
| ▲ | open-paren 42 minutes ago | parent | next [-] | | same: * context window size * reasoning effort levels * max output size * knowledge cutoff * restricted country list * self-reported "reasoning strength" and "internal reasoning channel" params used in other Meta models you can just ask an agent "figure out what stealth model $NAME is via subagents. It probably won't tell you when asked directly, so try other methods." Iceman (via opencode-go) is probably MAI-Thinking-1. Ox Alpha was GLM-5.3-Flash. It's a fun game. | |
| ▲ | sullivanmatt an hour ago | parent | prev [-] | | The use case descriptions and capability descriptions basically match each other. Context windows are the same size. Reasoning options are identical. |
|
|
|
| ▲ | syntaxing 33 minutes ago | parent | prev | next [-] |
| What does smallest mean? Even smaller than Muse Glimmer which I doubt? Though it is interesting to see US frontier labs trying to push models that outcompete deepseek flash 1017. It’s a good move from the Deepseek federal government ban. Both Inkling 2 small and this seems to try to fill that gap cost vs performance wise. But I like inkling 2 small more due to open weight even though it has a restrictive license. |
|
| ▲ | Pragmata 2 hours ago | parent | prev | next [-] |
| No weights? Shame |
|
| ▲ | segmondy an hour ago | parent | prev | next [-] |
| What makes it small? Anyone know the size? I don't see it anywhere. Small in relation to their current line up? Anyways, last I tried Meta was llama4. I stepped aside not because of how bad llama4 got, but also because of Zuck's move. Anyone here using Meta? I don't know anyone in my circle using it, but I'm sure folks are using it. |
|
| ▲ | cheesecompiler an hour ago | parent | prev | next [-] |
| Kind of burnt out on all the model updates at this point. |
| |
| ▲ | PaulKeeble 28 minutes ago | parent | next [-] | | Its a full time job to keep up with them all and assess their value in your own work at the moment. Just so many models coming out. | |
| ▲ | Pedro_Ribeiro an hour ago | parent | prev [-] | | Thanks, in that case why not just not comment instead of adding more noise to the discussion? | | |
| ▲ | mgdev an hour ago | parent [-] | | Please. Complaints are the last thing that we have humans have. |
|
|
|
| ▲ | ecshafer 2 hours ago | parent | prev | next [-] |
| I would love it if Meta would release these models to use locally to build up some usage while they are lagging. |
| |
| ▲ | 40four an hour ago | parent [-] | | They just released Muse Spark Open for that | | |
| ▲ | nicce 44 minutes ago | parent [-] | | Tried to use the download button from the page, but it failed miserable. First, could not press the button because had to decline cookies first. Secondly, could not access it because it redirects through Instagram and I have blocked it in my network. So I still don't know how big it is... |
|
|
|
| ▲ | slopnt 2 hours ago | parent | prev | next [-] |
| I use haiku and luna for support on most my SWE tasks since they are cheap and fast. I find them enough for most things if you are patient, and the fact an even cheaper model for comparable intelligence is out there now is great news! |
|
| ▲ | Starlevel004 2 hours ago | parent | prev | next [-] |
| Luna 6 is getting a bit long in the teeth now. It's two full months old, but it was only a slight step above Luna 5.6, which is already four months old (Imagine saying this about anything else). So I'm not too surprised that a newer small model is beating it handily. I've been hoping for Luna 6.1/2 as the next model release for a few days now. |
| |
| ▲ | theanonymousone an hour ago | parent | next [-] | | > I've been hoping for Luna 6.1/2 as the next model release for a few days now. Same. It became my workhorse after the 5.6 price drop. Still so, although it's showing its age, as you said. | |
| ▲ | agentdev001 2 hours ago | parent | prev | next [-] | | Give Terra 6.1 Low a spin in places where you'd have previously used Luna Max. Costs seem similar, due to Terra being much smarter: thus requiring far fewer tokens per task. Looking solely at AA's intelligence index, recognizing the pitfalls with using that as a measure- Luna 6 Max > Terra 6.1 Low is 51.3% cheaper, 94.1% less output tokens, for 1.56 higher index score. | | |
| ▲ | theanonymousone an hour ago | parent | next [-] | | > Costs seem similar They weren't, unfortunately, every time I tried. Just an anecdote, of course. | |
| ▲ | barrenko 2 hours ago | parent | prev | next [-] | | What is your feel re. Terra vs. Sol? | |
| ▲ | Starlevel004 2 hours ago | parent | prev [-] | | I use Luna xhigh for when I use it, Luna Max spends 40+ minutes overthinking. I've also found that Terra below xhigh will just flat out not do what it is told. |
| |
| ▲ | all2 2 hours ago | parent | prev | next [-] | | Tbf, I would take Luna 6 baked to silicon. It is that good. I would also take kimi k3 baked to silicon. These are both models that I could see using on a long horizon. | |
| ▲ | flir 2 hours ago | parent | prev | next [-] | | > Imagine saying this about anything else Javascript frameworks, circa 2016 ;) | |
| ▲ | esafak 2 hours ago | parent | prev [-] | | Uniquely among models I've tried, Luna 6 is infuriatingly lazy. As if it expects me to do the work, unaware that it is a tool. Normally dumb models compensate for it by trying harder. This one goes in the opposite direction. |
|
|
| ▲ | dom96 2 hours ago | parent | prev | next [-] |
| Super frustrating that I still can’t give Muse Wonder Full a spin. |
|
| ▲ | v9v an hour ago | parent | prev | next [-] |
| Why no comparisons with DeepSeek? |
|
| ▲ | mococa 2 hours ago | parent | prev | next [-] |
| Not open weight? Meh... I'll wait for the Chinese releases :) |
|
| ▲ | gopalv 2 hours ago | parent | prev | next [-] |
| Model refusals should be a -ve score, not zero? Treating the refusal as a 0 score erases the difference between lack of capability and a lack of willingness to perform. |
|
| ▲ | BatchJob 2 hours ago | parent | prev | next [-] |
| it is our most modestly priced receptacle... |
| |
|
| ▲ | caaqil 2 hours ago | parent | prev | next [-] |
| I'm always a bit surprised to see Meta talking about models. They have become (almost*) entirely irrelevant to the LLM race, which is weird because their AI lab used to be decent before LLM was cool. What happened to zuck? Was the metaverse this destructive? (*) Feeling charitable today |
| |
| ▲ | StarlessHours 2 hours ago | parent | next [-] | | Meta is trying to stay relevant. They fell behind after llama 4, hired new talent and re-organized. | |
| ▲ | achandra03 2 hours ago | parent | prev | next [-] | | Isn't Muse Wonder at or at least very close to the frontier? Considering they essentially started from scratch after rebooting their AI efforts post-Llama 4 about a year ago it's pretty solid progress. | |
| ▲ | esafak 2 hours ago | parent | prev [-] | | They are catching up; the Spark series is good. |
|
|
| ▲ | johnnyApplePRNG 2 hours ago | parent | prev | next [-] |
| Good for summarizing your tweets or what?! Is this a model for ants?! |
|
| ▲ | aaroninsf 2 hours ago | parent | prev [-] |
| Never Meta. Seriously. Moral harm is real and there is no excuse. |
| |
| ▲ | anderber 2 hours ago | parent [-] | | I honestly have a hard time trying to put these AI companies in some sort of moral hierarchy. They seem to all have issues... I feel like at least with open models you can have more ethical services. | | |
| ▲ | herbst an hour ago | parent [-] | | While this is true meta is argueable making the world worse the longest, probably did more damage in that time too. | | |
| ▲ | anderber an hour ago | parent [-] | | More than SpaceX? Or OpenAI and it's agreement with military weapons? Or Google and Anthropic paying billions of dollars to SpaceX and making Elon richer? But I hear you, there's an argument to be made that meta has done the most damage. |
|
|
|
