--- title: Introducing Muse Glimmer: An Open Agentic Model That... description: Meta Superintelligence Labs is releasing Muse Glimmer, a 30-billion-parameter open agentic model under Apache 2.0, designed to run locally on consumer hardware... canonical: https://daily.dev/posts/introducing-muse-glimmer-an-open-agentic-model-that-runs-on-your-device-tstjsayiu twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Introducing Muse Glimmer: An Open Agentic Model That Runs on Your Device | daily.dev og:description: Meta Superintelligence Labs is releasing Muse Glimmer, a 30-billion-parameter open agentic model under Apache 2.0, designed to run locally on consumer hardware... og:url: https://daily.dev/posts/introducing-muse-glimmer-an-open-agentic-model-that-runs-on-your-device-tstjsayiu og:image: https://api.daily.dev/og/posts/TStJSayiU.png og:image:alt: Introducing Muse Glimmer: An Open Agentic Model That Runs on Your Device og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Introducing Muse Glimmer: An Open Agentic Model That Runs on Your Device **[Hacker News](https://daily.dev/sources/hn)** · 6 min read · 2 upvotes · 0 comments ## Summary Meta Superintelligence Labs is releasing Muse Glimmer, a 30-billion-parameter open agentic model under Apache 2.0, designed to run locally on consumer hardware (Mac, PC with a single GPU). The model is optimized for always-on local agent workflows including function calling, coding, and LLM-as-a-judge evaluation. Key technical highlights include: logit distillation from a larger teacher model (Muse Spark), 4-bit quantization shrinking the model to under 20 GB (fitting within 24–32 GB GPU envelopes), and speculative decoding via a DFlash-based drafter for faster token generation. Muse Glimmer supports multimodal input, 100+ languages, failure recovery, and scaffold compatibility. Weights are available on Hugging Face, with integrations planned for llama.cpp, MLX, ExecuTorch, Ollama, LM Studio, vLLM, and SGLang. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is Meta's Muse Glimmer model and how big is it? Muse Glimmer is a 30-billion-parameter open agentic model from Meta Superintelligence Labs, released under the Apache 2.0 license. It is optimized for always-on local agent workflows such as function calling, local coding, and LLM-as-a-judge evaluation, and is designed to run on a Mac or PC with a single consumer GPU. _developers evaluating local agent models can track releases like Muse Glimmer through daily.dev._ ### How does Muse Glimmer fit a 30B parameter model onto a consumer GPU? Muse Glimmer uses quantization to compress weights to roughly 4-bit precision, shrinking the model from over 55GB at full precision to under 20GB. This leaves headroom within a 24GB or 32GB memory envelope for the KV cache, a perception encoder for image understanding, and a speculative decoding drafter to run simultaneously. _daily.dev helps developers weighing hardware constraints for local model deployment stay current on techniques like this._ ### What is the DFlash drafter model used for speculative decoding in Muse Glimmer? DFlash is a lightweight companion network shipped with Muse Glimmer that proposes entire blocks of tokens at once, which the main model then verifies in parallel, accepting correct tokens and correcting wrong ones. This speculative decoding approach speeds up generation significantly compared to standard token-by-token decoding while preserving identical output quality. _engineers optimizing inference speed for local agents can follow developments like DFlash via daily.dev._ ## Community take How the wider developer community reacted, aggregated from 1 discussion and 113 comments across hackernews (as of 2026-09-13). **TL;DR:** Developers are cautiously optimistic about Muse Glimmer's open-weights release and competitive benchmarks, but raise real concerns about hardware requirements, inference speed on consumer GPUs, and whether it can hold its own against the imminent Qwen3.8 27B. **Sentiment:** 45% positive · 35% mixed · 20% skeptical **The case for** - Apache 2.0 licensing is more permissive than previous Llama licenses, setting a positive precedent for the industry. - Pre-quantized 4-bit weights (~20 GB) and a bundled DFlash speculative-decoding drafter make local deployment more practical out of the box. - Benchmark comparisons against Gemma 4 and Qwen3.6 look competitive, especially on tool-calling tasks. - Privacy and cost-control benefits of running a capable model fully locally are meaningful for many developers. - Multi-token prediction (MTP) helps dense models approach MoE speeds without sacrificing intelligence. **The pushback** - Effective memory footprint including KV cache, perception encoder, and drafter likely exhausts a 24 GB GPU at long context, which is exactly the agentic workload it targets. - Dense architecture means slow decode on DDR5/unified-memory systems (~15 tok/s on a DGX Spark), unlike MoE alternatives. - Benchmarks compare against Qwen3.6 27B (4 months old), and Qwen3.8 27B is expected to surpass it imminently. - Hardware capable of running it comfortably (64 GB Mac or 24 GB GPU laptop) costs $3,500–$4,000+, limiting accessibility. - "Minimal to no degradation" from 4-bit quantization on tool-call schema adherence is an unverified strong claim with no per-quant breakdown provided. - Energy costs of local inference likely mean users never break even versus cheap API providers. **By community** - hackernews (mixed): Broadly welcoming of the open-weights release and Apache 2.0 license, but skeptical about real-world hardware requirements, inference speed, benchmark honesty, and whether Qwen3.8 will immediately obsolete it. **Hottest debate:** Whether the hardware cost and inference speed of a dense 30B model make local deployment practical compared to cheap cloud API alternatives or MoE models of similar size. **Open questions** - How does Muse Glimmer perform at 64k+ context lengths on a 24 GB GPU once KV cache, encoder, and drafter are all resident? - Will Qwen3.8 27B surpass Glimmer-30B on most benchmarks when it releases? - Does the 4-bit quantization cause measurable degradation in tool-call schema adherence (malformed JSON, wrong enum values) that benchmark averages don't capture? - Is the open-weight version of Muse Spark 1.2 identical to what Meta serves internally? **Highlights** > The memory math is the part I keep re-reading. 4-bit gets the LM under 20GB, but they're explicitly budgeting the KV cache, the perception encoder, and the DFlash drafter into the same 24GB envelope. On a 4090 that leaves maybe 3-4GB of KV once the drafter and encoder are resident, so long agent traces are going to spill or truncate — which is exactly the workload this model is trained for. I'd like to see the K-Quant-17GB numbers reported at 64k+ context, not just conversational length. "Minimal to no degradation on agentic tasks" from quantization is also a strong claim. In my experience 4-bit shows up first in tool-call schema adherence — malformed JSON args, wrong enum values — before it moves benchmark averages. Does the report break down tau-Bench / MCP-Atlas per quant level? > — [jkwang on hackernews](https://news.ycombinator.com/item?id=49242100) > Wow, Meta is back (at least for now)! I like this class of model. Multi-token prediction makes it viable to run dense models at not-too-far-off speeds as MoE models with much better intelligence. The submission’s title (open weights 30B local coding model) is luckily wrong: This is meant to be a general agentic model. It even comes pre-quantized and with a MTP/drafter model. Looking good! Let’s hope they aren’t dishonest with the benchmarks this time … > — [solarkraft on hackernews](https://news.ycombinator.com/item?id=49242071) > It is interesting but it does look like a careful distillation of (Spark and) biggers open-weight models. The progress compared to Qwen3.6 27B is good, not that impressive, it's a 4 months old model. (kuto to them to compare to 27B dense and not 35B MoE, it's more fair to do so). It is very probable that Qwen3.8 27B will crush Glimmer-30B on most benchmarks. > — [\_ache\_ on hackernews](https://news.ycombinator.com/item?id=49242292) > This model I think will be too slow for that on Spark, even at 4 bit quant. It's a dense model, not MoE like e.g. Qwen 35b. On a Spark it will be memory bandwidth limited (unlike Gemma 4.) I haven't tried yet (working on it) but back of the napkin estimate puts it at around 15tok/s even after converting to NVFP4. Prefill would be much higher though. That 15tok/sec is pretty typical for dense models of this size. NVFP4 Q/K/V/O and MLP projections: ~13 GB/token BF16 attention gates: ~3 GB/token BF16 LM head: ~2.5 GB/token Total: ~18.9 GB/token At 273 GB/s, that gives a bandwidth-only ceiling of about 14.5 tok/s; actual performance would be lower. > — [cmrdporcupine on hackernews · 1 comments](https://news.ycombinator.com/item?id=49242179) > Some interesting findings from the chat template designs: 1. The template name is Onyx ATEM as found in the tool call exception message 2. It appears to be following a harmony-style chat template. But the tool use seems to be a xml like : / / 3. atem: a internal joke of meta in reverse? https://huggingface.co/meta-models/Muse-Glimmer-30B/blob/mai... > — [polymorph1sm on hackernews · 1 comments](https://news.ycombinator.com/item?id=49242626) **Source threads** - [hackernews](https://news.ycombinator.com/item?id=49241679) · 268 points · 113 comments ## Similar posts on daily.dev - [Fine-tune Meta AI's Muse Glimmer 100% locally](https://daily.dev/posts/fine-tune-meta-ai-s-muse-glimmer-100-locally-nvrlxatlc) · Daily Dose of Data Science \| Avi Chawla \| Substack · 0 upvotes · 0 comments - [Introducing Muse Spark 1.1](https://daily.dev/posts/introducing-muse-spark-1-1-6ixvwquql) · AI at Meta · 15 upvotes · 1 comments --- Tags: [#data-science](https://daily.dev/tags/data-science), [#llm](https://daily.dev/tags/llm), [#ai-agents](https://daily.dev/tags/ai-agents), [#ai-inference](https://daily.dev/tags/ai-inference) [View this post on daily.dev](https://daily.dev/posts/introducing-muse-glimmer-an-open-agentic-model-that-runs-on-your-device-tstjsayiu) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Introducing Muse Glimmer: An Open Agentic Model That Runs on Your Device","url":"https://daily.dev/posts/introducing-muse-glimmer-an-open-agentic-model-that-runs-on-your-device-tstjsayiu","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/introducing-muse-glimmer-an-open-agentic-model-that-runs-on-your-device-tstjsayiu"},"datePublished":"2026-08-10T11:04:24.422Z","dateModified":"2026-09-13T21:21:58.372Z","description":"Meta Superintelligence Labs is releasing Muse Glimmer, a 30-billion-parameter open agentic model under Apache 2.0, designed to run locally on consumer hardware...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/28c9682d6759c428e43c57d1c6816158?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/28c9682d6759c428e43c57d1c6816158?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Hacker News","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Hacker News","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/hn","url":"https://daily.dev/sources/hn"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/introducing-muse-glimmer-an-open-agentic-model-that-runs-on-your-device-tstjsayiu","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"data-science,llm,ai-agents,ai-inference","timeRequired":"PT6M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Hacker News","item":"https://daily.dev/sources/hn"},{"@type":"ListItem","position":3,"name":"Introducing Muse Glimmer: An Open Agentic Model That Runs on Your Device"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/introducing-muse-glimmer-an-open-agentic-model-that-runs-on-your-device-tstjsayiu#faq","mainEntity":[{"@type":"Question","name":"What is Meta's Muse Glimmer model and how big is it?","acceptedAnswer":{"@type":"Answer","text":"Muse Glimmer is a 30-billion-parameter open agentic model from Meta Superintelligence Labs, released under the Apache 2.0 license. It is optimized for always-on local agent workflows such as function calling, local coding, and LLM-as-a-judge evaluation, and is designed to run on a Mac or PC with a single consumer GPU. developers evaluating local agent models can track releases like Muse Glimmer through daily.dev."}},{"@type":"Question","name":"How does Muse Glimmer fit a 30B parameter model onto a consumer GPU?","acceptedAnswer":{"@type":"Answer","text":"Muse Glimmer uses quantization to compress weights to roughly 4-bit precision, shrinking the model from over 55GB at full precision to under 20GB. This leaves headroom within a 24GB or 32GB memory envelope for the KV cache, a perception encoder for image understanding, and a speculative decoding drafter to run simultaneously. daily.dev helps developers weighing hardware constraints for local model deployment stay current on techniques like this."}},{"@type":"Question","name":"What is the DFlash drafter model used for speculative decoding in Muse Glimmer?","acceptedAnswer":{"@type":"Answer","text":"DFlash is a lightweight companion network shipped with Muse Glimmer that proposes entire blocks of tokens at once, which the main model then verifies in parallel, accepting correct tokens and correcting wrong ones. This speculative decoding approach speeds up generation significantly compared to standard token-by-token decoding while preserving identical output quality. engineers optimizing inference speed for local agents can follow developments like DFlash via daily.dev."}}]} ```