muse glimmer has better taste than expected on voxel worlds. both one-shot, generated using my GX10 structure isn't perfect and it's less convincing than ds4 flash or qwen 3.6 but it's good enough!

Uncensored Meta Muse Glimmer 30B model run locally 10.7 GB on 5090. - Claim on the card: 0/300 measured refusals. - 131k context. - image understanding Agent out. - compact vision projector + DFlash drafter included - claimed 0 true refusals on a 450-prompt harmful suite -

You can now fine-tune Meta Muse Glimmer 30B for free! 🔥 Our free notebook also supports GRPO RL training. Unsloth trains Muse Glimmer 1.5× faster with 50% less VRAM vs FA2 setups. Train locally with 24GB VRAM. Guide: Notebooks:

I shipped a fine-tuning tutorial for Muse Glimmer 30B on AI2's MolmoWeb dataset with TRL > but merve, this model is sota on ScreenSpot-Pro? yes, but it doesn't work well with ambiguous prompts of MolmoWeb that are close to how you interact with computer, e.g. "jump to nutrition facts" I compared model fine-tuned on MolmoWeb format against base model zero-shot outputs converted to MolmoWeb format (coordinates on 100) I found that base model has 13% click accuracy + 35.0% within 5% diagonal while fine-tuned model has 41% + 68% within 5% diagonal

free Kaggle dual T4s GPUs Running Meta’s Muse Glimmer 30B Q6_K_XL. - 130,000 token context - 26.5 GB VRAM - Prefill: 265 t/s - Decode: 9 t/s - Cost: $0 Aggressive 16:1 GQA. Unquantized KV cache for 130k tokens is only 2 GB.

Today we're also opening the weights for Muse Glimmer, a great 30B parameter dense model that can run locally. Soon we'll also release the weights for Muse Spark 1.2, our latest foundation model. Meta is a strong supporter of open source and I'm proud of these releases. Congrats to @alexandr_wang and the MSL team for all your great work on these models.

Okay, well since I can't get access to Muse Spark through the Meta API yet, I'm going to benchmark Muse Glimmer with @VulcanBench 🖖 This will be the first model I test that has a cost of $0, which I like a lot! If you don't know about Muse Glimmer, it was just announced by @finkd this week. Major step forward for American models, and something that I should be able to run locally on my four year old M1 Mac Studio, which is pretty darn cool.

Today’s video tries to answer the question: Which model that can fit on DGX Spark can perform autoresearch the best? The contenders: 1. Ornith 1.5 35B A3B 2. Nemotron 3.5 Lightning 3. Muse Glimmer 30B 4. Qwen3.8-27B Interesting results and insights from this one! Which Local LLM Performs Autoresearch the Best? (Ornith, Qwen, Nemotron, Muse Glimmer)