community pool / model answer

Muse-Glimmer-30B-GGUF

chat · meta-models/Muse-Glimmer-30B-GGUF

28.2 tok/s · model median · 15 runs in the model record.

Reference rig	Quantization	Median result	Basis	TTFT (ms)	Peak VRAM	Max context	Engines
Arc Pro B70 32GB	4-bit	28.8 tok/s	measured	210	22 GB	2,048	llama.cpp
Radeon AI Pro R9700 32GB	4-bit	82.5 tok/s	reported	259	23.7 GB	32,768	llama.cpp
RTX 3080 12GB	4-bit	6 tok/s	reported	3,840	8.8 GB	16,384	llama.cpp
RTX PRO 6000 Blackwell 96GB	4-bit	167.5 tok/s	reported	561	24.3 GB	131,072	llama.cpp
Ryzen AI Max 395 64GB	4-bit	12.9 tok/s	measured	4,558	16.4 GB	8,192	llama.cpp

Context tested

The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.