Benchmark result

unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL on NVIDIA GeForce RTX 3080 Ti +1 more — 68.1 tok/s

Measured with llama.cpp b11149-d2e54583c on September 23, 2026.

ShareReddit X

What the model built

Open full screen →

The coding scenario asks for a playable game in a single HTML file. This is exactly what the model returned, unedited.

How this run compares

unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL on other hardware

34.3tok/s82.2Advanced Micro Devices, Inc. [AMD/ATI] Device 7551 · Q4_K

Reproduce this run

toolllama.cpp b11149-d2e54583c

modelunsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL (Q4_K)

context65,536

samplingtemperature 1 · top_p 0.949999988079071 · top_k 64 · min_p 0 · presence_penalty 0 · repeat_penalty 1

clientv0.4.80+98

Set those in llama.cpp, then:

llm-benchmark benchmark --model "unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL" --tool "llama.cpp"

The client prompts for context length and thinking mode, and for the KV cache dtype on Unsloth Studio. Sampling is left at the model default — the values above are what the tool reported using, not overrides the client sent.