RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can I run Muse Glimmer 30B locally?

Muse Glimmer 30B by meta-models has 30B parameters. At 4-bit with an 8K context it needs about 17.9 GB of GPU memory. It fits on 56 of 65 GPUs and Macs we track. See all models that fit in 24 GB of VRAM.

17.9 GBQ4_K_M · 8K context
RTX 3090smallest consumer GPU that fits
M4 32GBsmallest Mac that fits
128Kmax context
GQA + sliding windowattention type

Memory needed by quant and context length

GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use real file sizes from unsloth/Muse-Glimmer-30B-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx8K ctx32K ctx128K ctx
2-bit (Q2_K)rough11.3 GB11.4 GB11.7 GB12.9 GB
3-bit (Q3_K_M)usable14.5 GB14.5 GB14.8 GB16.0 GB
4-bit (Q4_K_M)good, most popular17.9 GB17.9 GB18.2 GB19.4 GB
5-bit (Q5_K_M) •very good18.8 GB18.8 GB19.1 GB20.4 GB
6-bit (Q6_K)near lossless23.7 GB23.7 GB24.0 GB25.2 GB
8-bit (Q8_0) •effectively lossless28.5 GB28.5 GB28.8 GB30.1 GB
16-bit (F16/BF16) •original weights52.8 GB52.9 GB53.2 GB54.4 GB

Which GPUs and Macs run Muse Glimmer 30B?

Best quality quant that fits at 8K context. "Tight" = within 10% of usable memory. Speed is a ceiling from memory bandwidth (real speed is typically 50–80% of it).

HardwareMemoryFits?Best quantNeedsMax tok/s (est.)
RTX 3060 12GB12 GBTightQ2_K11.4 GB32details
RTX 4060 Ti 16GB16 GBFitsQ2_K11.4 GB26details
RTX 309024 GBFitsQ5_K_M18.8 GB49details
RTX 409024 GBFitsQ5_K_M18.8 GB53details
RTX 509032 GBFitsQ6_K23.7 GB73details
RTX 5070 Ti16 GBFitsQ2_K11.4 GB80details
RX 7900 XTX24 GBFitsQ5_K_M18.8 GB50details
Mac M4 Pro 48GB48 GBFitsQ8_028.5 GB9.2details
Mac M4 Max 128GB128 GBFitsF1652.9 GB9.8details
Mac M3 Ultra 512GB512 GBFitsF1652.9 GB15details
DGX Spark128 GBFitsF1652.9 GB4.9details
RTX 407012 GBTightQ2_K11.4 GB45details
RTX 5060 Ti 16GB16 GBFitsQ2_K11.4 GB40details
RTX 40608 GBNo———details
Arc B58012 GBTightQ2_K11.4 GB41details
RTX 3060 Ti8 GBNo———details
RTX 30708 GBNo———details
RTX 3080 10GB10 GBNo———details
RTX 3080 12GB12 GBTightQ2_K11.4 GB82details
RTX 3090 Ti24 GBFitsQ5_K_M18.8 GB53details
RTX 4060 Ti 8GB8 GBNo———details
RTX 4070 Super12 GBTightQ2_K11.4 GB45details
RTX 4070 Ti12 GBTightQ2_K11.4 GB45details
RTX 4070 Ti Super16 GBFitsQ2_K11.4 GB60details
RTX 408016 GBFitsQ2_K11.4 GB64details
RTX 4080 Super16 GBFitsQ2_K11.4 GB66details
RTX 50608 GBNo———details
RTX 507012 GBTightQ2_K11.4 GB60details
RTX 508016 GBFitsQ2_K11.4 GB86details
2x RTX 30902×24 GBFitsQ8_028.5 GB32details
2x RTX 40902×24 GBFitsQ8_028.5 GB34details
2x RTX 50902×32 GBFitsF1652.9 GB32details
RTX A600048 GBFitsQ8_028.5 GB26details
RTX 6000 Ada48 GBFitsQ8_028.5 GB32details
RTX PRO 6000 Blackwell96 GBFitsF1652.9 GB32details
L40S48 GBFitsQ8_028.5 GB29details
A100 40GB40 GBFitsQ8_028.5 GB53details
A100 80GB80 GBFitsF1652.9 GB37details
H100 PCIe80 GBFitsF1652.9 GB36details
H100 SXM80 GBFitsF1652.9 GB60details
H200141 GBFitsF1652.9 GB86details
RX 76008 GBNo———details
RX 7700 XT12 GBTightQ2_K11.4 GB39details
RX 7800 XT16 GBFitsQ2_K11.4 GB56details
RX 7900 GRE16 GBFitsQ2_K11.4 GB52details
RX 7900 XT20 GBFitsQ3_K_M14.5 GB55details
RX 907016 GBFitsQ2_K11.4 GB57details
RX 9070 XT16 GBFitsQ2_K11.4 GB57details
Ryzen AI Max+ 395 128GB128 GBFitsF1652.9 GB4.6details
Arc A770 16GB16 GBFitsQ2_K11.4 GB50details
Mac M1 16GB16 GBNo———details
Mac M2 24GB24 GBFitsQ2_K11.4 GB9.0details
Mac M3 24GB24 GBFitsQ2_K11.4 GB9.0details
Mac M4 16GB16 GBNo———details
Mac M4 32GB32 GBFitsQ5_K_M18.8 GB6.3details
Mac M1 Pro 32GB32 GBFitsQ5_K_M18.8 GB10details
Mac M3 Pro 36GB36 GBFitsQ5_K_M18.8 GB7.8details
Mac M4 Pro 24GB24 GBFitsQ2_K11.4 GB24details
Mac M4 Pro 64GB64 GBFitsQ8_028.5 GB9.2details
Mac M1 Max 64GB64 GBFitsQ8_028.5 GB14details
Mac M2 Max 96GB96 GBFitsF1652.9 GB7.2details
Mac M3 Max 128GB128 GBFitsF1652.9 GB7.2details
Mac M4 Max 64GB64 GBFitsQ8_028.5 GB18details
Mac M2 Ultra 192GB192 GBFitsF1652.9 GB14details
Mac M3 Ultra 256GB256 GBFitsF1652.9 GB15details

Prefer an API?

Meta: Muse Glimmer 30B is on OpenRouter at $0.35 per 1M input tokens and $1.50 per 1M output tokens. Price details and history →

Questions

How much VRAM does Muse Glimmer 30B need?

About 17.9 GB at 4-bit (Q4_K_M) with an 8K context: 17.0 GB of weights (estimated), 0.2 GB of KV cache and 0.8 GB of runtime overhead. At 8-bit it needs about 28.5 GB.

What is the smallest GPU that runs Muse Glimmer 30B?

At Q4_K_M and 8K context, the smallest single consumer GPU in our table that fits it is the RTX 3090 (24 GB).

Can I run Muse Glimmer 30B on a Mac?

Yes. At Q4_K_M with 8K context it fits on a Mac M4 32GB or larger, using the default macOS GPU memory limit.

Similar models

Same family or a similar parameter count, with the VRAM each needs at Q4_K_M and 8K context.

Muse Spark Open 816B A42B

822B · 470 GB at 4-bit

Hy Memory 30B A3B

30B · 18.5 GB at 4-bit

granite 4.2 30b

29B · 19.1 GB at 4-bit

North Mini Code 1.0

30B · 19.3 GB at 4-bit

Qwen3 Coder 30B A3B Instruct

31B · 18.7 GB at 4-bit

GLM 4.7 Flash

31B · 18.2 GB at 4-bit

Model facts

Hugging Face repometa-models/Muse-Glimmer-30B
Parameters29,776,626,688 (30B)
Layers / hidden size52 / 6656
AttentionGQA + sliding window; 32 heads, 2 KV heads, head dim 128; sliding window 2048
KV cache per token (FP16)52.0 KiB (sliding layers capped at the window)
Max context131,072 tokens
Licenseapache-2.0
Released on HFAug 9, 2026
Downloads (30d) / likes148,951 / 2,157

Sources: Hugging Face model API · config.json (meta-models/Muse-Glimmer-30B) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.