Can Muse Glimmer 30B run on the A100 40GB?
Yes. At 8K context the best fit is Q8_0, using about 28.5 GB of the 38.0 GB usable on the A100 40GB. Estimated speed ceiling: ~53 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | 128K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|---|
| 2-bit (Q2_K) | 11.3 GB Fits | 11.4 GB Fits | 11.7 GB Fits | 12.9 GB Fits | 139 |
| 3-bit (Q3_K_M) | 14.5 GB Fits | 14.5 GB Fits | 14.8 GB Fits | 16.0 GB Fits | 107 |
| 4-bit (Q4_K_M) | 17.9 GB Fits | 17.9 GB Fits | 18.2 GB Fits | 19.4 GB Fits | 85 |
| 5-bit (Q5_K_M) • | 18.8 GB Fits | 18.8 GB Fits | 19.1 GB Fits | 20.4 GB Fits | 81 |
| 6-bit (Q6_K) | 23.7 GB Fits | 23.7 GB Fits | 24.0 GB Fits | 25.2 GB Fits | 64 |
| 8-bit (Q8_0) • | 28.5 GB Fits | 28.5 GB Fits | 28.8 GB Fits | 30.1 GB Fits | 53 |
| 16-bit (F16/BF16) • | 52.8 GB No | 52.9 GB No | 53.2 GB No | 54.4 GB No | — |
• real GGUF file size from unsloth/Muse-Glimmer-30B-GGUF. Speed is a bandwidth ceiling (1555 GB/s), not a benchmark. Methodology.
More
Muse Glimmer 30B on every GPU and Mac · Every model that runs on the A100 40GB · Other models that fit 48 GB VRAM · Try other settings in the calculator
Data: Hugging Face Hub (meta-models/Muse-Glimmer-30B), NVIDIA specs. Data as of 2026-11-17.