Measured model
Measured on a Mac Studio (M5 Max, 36GB)muse-glimmer:30b
Meta
On a Mac Studio (M5 Max, 36GB), muse-glimmer:30b answers a simple question in about 14.3 s, gets 85% of the Japanese knowledge questions right, 88% of the math problems, and uses 17.6GB of memory.
These numbers are from before the method changed on Oct 2, 2026 (see how it was measured).
Results
| Measure | Result | Among measured models | Note |
|---|---|---|---|
| Reply time | about 14.3 s | 28 of 30 | Until one simple common-sense question is answered (median of 50) |
| Knowledge (high school to university) | 85% | 4 of 30 | 200 questions, ±6 |
| Math (word problems) | 88% | 3 of 30 | 100 questions, ±8 |
| Memory used | 17.6GB | 20 of 30 | |
| EN→JA translation | 36 pts | 1 of 28 | out of 100 (chrF), about 19 s a sentence |
| JA→EN translation | 60 pts | 1 of 28 | out of 100 (chrF), about 21.4 s a sentence |
"Among measured models" is the place among the 30 models measured on this site (lower is better for time and memory).
Try it on your Mac
- Install the Mac version of Ollama from ollama.com and start it.
- Type this in Terminal. The first time, the model downloads; then you can chat with it.
ollama run muse-glimmer:30b
It used 17.6GB here. To use it comfortably alongside other apps, a Mac with 36GB of memory or more is a good guide (the model then takes about half the memory).
About the model
- Maker
- Meta
- Size
- 27.9B parameters
- Quantization
- Q4_K_M
- Thinks before answering
- Yes
- Longest input at once
- 131,072 tokens
- Released (Hugging Face)
- Aug 9, 2026
- Generation speed
- 25 tok/s
- Start-up (loading)
- about 6.1 s
- Measured on
- Sep 26, 2026
- Ollama
- 0.34.3
- Hugging Face
- meta-models/Muse-Glimmer-30B
- Series
- Muse
Models of similar memory
- qwen3.8:27b (18.2GB of memory, knowledge 85%, reply about 6.9 s)
- gemma3:27b (18.4GB of memory, knowledge 74%, reply about 0.4 s)
- gemma4:26b (18.7GB of memory, knowledge 82%, reply about 2.9 s)
- Devstral-Small-2-24B-Instruct-2512 (16.3GB of memory, knowledge 66%, reply about 0.2 s)
- Mistral-Small-3.1-24B-Instruct-2503 (16.3GB of memory, knowledge 72%, reply about 0.2 s)
Compare the main models on the text page. The series page lists every model measured.