AMD-Pro-BenchCommunity

13,812 benchmarks in 32 GPU combinations documented

Muse-Glimmer-30B on AMD GPUs

45 results · 5 sources · 160 measurements

Explore all results

Grouped by reported model identity. Exact weights, quantizations and software can differ. How results are reviewed

Documented results

24 selected tests
View
ReportCustom evaluation		Muse-Glimmer-30BQ4_K_MModel ref. ↗Conflicting evidenceHumanEval100 problems
97 %Reported pass@1 · n = 100
—
27.51 tok/sGeneration · mean · per request
1 × W7800Single GPU · 48 GB VRAM/GPU
Configuration
Memory
48 GB VRAM/GPU
OS
Ubuntu 24.04 LTS
Engine
Ollama
Backend
ROCm
Harness
BoJl unified evaluation harness (2026-08)
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
ReportLocalMaxxing		Muse-Glimmer-30BQ4_K_MModel ref. ↗Token speed—
122.9 tok/sPrefill · median · unknown
27.7 tok/sGeneration · median · unknown
1 × Ryzen AI Max+ 395Single GPU · Unified memory
Configuration
Memory
Unified memory
OS
Ubuntu 24.04
Engine
llama.cpp
Backend
vulkan
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
123 / 256 tokens
Context
4096
Concurrency
Not reported
Date
2026-08-12
Full record →
View
ReportLocalMaxxing		Muse-Glimmer-30BROCMFP4_FASTModel ref. ↗Token speed—
754.03 tok/sPrefill · median · unknown
42.98 tok/sGeneration · median · unknown
3 × R9700Multi-GPU · split unknown · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Arch Linux
Engine
llama.cpp
Backend
rocm
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
1116 / 256 tokens
Context
1372
Concurrency
Not reported
Date
2026-08-11
Full record →
View
ReportGeneric LLM		Muse-Glimmer-30BUD-Q2-K-XLModel ref. ↗Token speed—
~308 tok/sPrefill · per request
~20 tok/sGeneration · per request
1 × RX 7600 XTSingle GPU · 16 GB VRAM/GPU
Configuration
Memory
16 GB VRAM/GPU
OS
Debian Sid (guest); Debian Trixie (host)
Engine
llama.cpp
Backend
ROCm / HIP
Harness
Author coding request timing
Prompt / output
? / ? tokens
Context
62144
Concurrency
Not reported
Date
Not reported
Full record →
View
ReportGeneric LLM		Muse-Glimmer-30BQuant not reportedModel ref. ↗Token speed——
53 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Windows 11 Pro 25H2
Engine
llama.cpp
Backend
Vulkan
Harness
Not reported
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-08-10
Full record →
View
ReportCustom evaluation		Muse-Glimmer-30BQ4_K_MModel ref. ↗Conflicting evidenceToken speedsynth-bulk-dispatcher-impl——
26.88 tok/sGeneration · mean · per request
1 × W7800Single GPU · 48 GB VRAM/GPU
Configuration
Memory
48 GB VRAM/GPU
OS
Ubuntu 24.04 LTS
Engine
Ollama
Backend
ROCm
Harness
BoJl unified evaluation harness (2026-08)
Prompt / output
? / 4096 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
ReportLocalMaxxing		Muse-Glimmer-30BQ4_K_MModel ref. ↗Token speed—
260.2 tok/sPrefill · median · unknown
12.9 tok/sGeneration · median · unknown
1 × Ryzen AI Max+ 395Single GPU · Unified memory
Configuration
Memory
Unified memory
OS
Ubuntu 24.04
Engine
llama.cpp
Backend
vulkan
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
1210 / 512 tokens
Context
8192
Concurrency
Not reported
Date
2026-08-12
Full record →
View
ReportLocalMaxxing		Muse-Glimmer-30BUD-Q8_K_XLModel ref. ↗Token speed—
856.13 tok/sPrefill · median · unknown
34.06 tok/sGeneration · median · unknown
3 × R9700Multi-GPU · split unknown · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Arch Linux
Engine
llama.cpp
Backend
rocm
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
1116 / 256 tokens
Context
1372
Concurrency
Not reported
Date
2026-08-11
Full record →
View
ReportCustom evaluation		Muse-Glimmer-30BQ4_K_MModel ref. ↗Conflicting evidenceToken speedsynth-horner-mpmath-derivative——
27.17 tok/sGeneration · mean · per request
1 × W7800Single GPU · 48 GB VRAM/GPU
Configuration
Memory
48 GB VRAM/GPU
OS
Ubuntu 24.04 LTS
Engine
Ollama
Backend
ROCm
Harness
BoJl unified evaluation harness (2026-08)
Prompt / output
? / 2012 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
ReportLocalMaxxing		Muse-Glimmer-30BQ4_K_MModel ref. ↗Token speed—
268.5 tok/sPrefill · median · unknown
12.9 tok/sGeneration · median · unknown
1 × Ryzen AI Max+ 395Single GPU · Unified memory
Configuration
Memory
Unified memory
OS
Ubuntu 24.04
Engine
llama.cpp
Backend
vulkan
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
1210 / 512 tokens
Context
8192
Concurrency
Not reported
Date
2026-08-12
Full record →
View
ReportLocalMaxxing		Muse-Glimmer-30BUD-Q8_K_XLModel ref. ↗Token speed—
683.69 tok/sPrefill · median · unknown
26.69 tok/sGeneration · median · unknown
3 × R9700Multi-GPU · split unknown · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Arch Linux
Engine
llama.cpp
Backend
rocm
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
578 / 238 tokens
Context
816
Concurrency
Not reported
Date
2026-08-11
Full record →
View
ReportCustom evaluation		Muse-Glimmer-30BQ4_K_MModel ref. ↗Conflicting evidenceToken speedsynth-invoke-sign-artifacts-ps——
27.11 tok/sGeneration · mean · per request
1 × W7800Single GPU · 48 GB VRAM/GPU
Configuration
Memory
48 GB VRAM/GPU
OS
Ubuntu 24.04 LTS
Engine
Ollama
Backend
ROCm
Harness
BoJl unified evaluation harness (2026-08)
Prompt / output
? / 2532 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
ReportLocalMaxxing		Muse-Glimmer-30BQ4_K_MModel ref. ↗Token speed—
124.5 tok/sPrefill · median · unknown
28.2 tok/sGeneration · median · unknown
1 × Ryzen AI Max+ 395Single GPU · Unified memory
Configuration
Memory
Unified memory
OS
Ubuntu 24.04
Engine
llama.cpp
Backend
vulkan
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
123 / 256 tokens
Context
4096
Concurrency
Not reported
Date
2026-08-12
Full record →
View
ReportLocalMaxxing		Muse-Glimmer-30BUD-Q8_K_XLModel ref. ↗Token speed—
991.06 tok/sPrefill · median · unknown
29.66 tok/sGeneration · median · unknown
3 × R9700Multi-GPU · split unknown · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Arch Linux
Engine
llama.cpp
Backend
rocm
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
1411 / 512 tokens
Context
1923
Concurrency
Not reported
Date
2026-08-11
Full record →
View
ReportCustom evaluation		Muse-Glimmer-30BQ4_K_MModel ref. ↗Conflicting evidenceToken speedsynth-legacy-rename-toolchain——
27.29 tok/sGeneration · mean · per request
1 × W7800Single GPU · 48 GB VRAM/GPU
Configuration
Memory
48 GB VRAM/GPU
OS
Ubuntu 24.04 LTS
Engine
Ollama
Backend
ROCm
Harness
BoJl unified evaluation harness (2026-08)
Prompt / output
? / 1602 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
ReportLocalMaxxing		Muse-Glimmer-30BQ4_K_MModel ref. ↗Token speed—
255.9 tok/sPrefill · median · unknown
12.8 tok/sGeneration · median · unknown
1 × Ryzen AI Max+ 395Single GPU · Unified memory
Configuration
Memory
Unified memory
OS
Ubuntu 24.04
Engine
llama.cpp
Backend
vulkan
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
1210 / 512 tokens
Context
8192
Concurrency
Not reported
Date
2026-08-12
Full record →
View
ReportLocalMaxxing		Muse-Glimmer-30BROCMFP4_FASTModel ref. ↗Token speed—
1,116.81 tok/sPrefill · median · unknown
62.55 tok/sGeneration · median · unknown
3 × R9700Multi-GPU · split unknown · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Arch Linux
Engine
llama.cpp
Backend
rocm
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
1411 / 512 tokens
Context
1923
Concurrency
Not reported
Date
2026-08-11
Full record →
View
ReportCustom evaluation		Muse-Glimmer-30BQ4_K_MModel ref. ↗Conflicting evidenceToken speedsynth-nuget-multitarget-design——
27.52 tok/sGeneration · mean · per request
1 × W7800Single GPU · 48 GB VRAM/GPU
Configuration
Memory
48 GB VRAM/GPU
OS
Ubuntu 24.04 LTS
Engine
Ollama
Backend
ROCm
Harness
BoJl unified evaluation harness (2026-08)
Prompt / output
? / 1024 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
ReportLocalMaxxing		Muse-Glimmer-30BQ4_K_MModel ref. ↗Token speed—
262.5 tok/sPrefill · median · unknown
12.7 tok/sGeneration · median · unknown
1 × Ryzen AI Max+ 395Single GPU · Unified memory
Configuration
Memory
Unified memory
OS
Ubuntu 24.04
Engine
llama.cpp
Backend
vulkan
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
1210 / 512 tokens
Context
8192
Concurrency
Not reported
Date
2026-08-12
Full record →
View
ReportLocalMaxxing		Muse-Glimmer-30BUD-Q8_K_XLModel ref. ↗Token speed—
787.46 tok/sPrefill · unknown
47.77 tok/sGeneration · unknown
3 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Arch Linux
Engine
llama.cpp
Backend
rocm
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
578 / 256 tokens
Context
834
Concurrency
Not reported
Date
2026-08-16
Full record →
View
ReportCustom evaluation		Muse-Glimmer-30BQ4_K_MModel ref. ↗Conflicting evidenceToken speedsynth-parser-nunit-tests——
27.04 tok/sGeneration · mean · per request
1 × W7800Single GPU · 48 GB VRAM/GPU
Configuration
Memory
48 GB VRAM/GPU
OS
Ubuntu 24.04 LTS
Engine
Ollama
Backend
ROCm
Harness
BoJl unified evaluation harness (2026-08)
Prompt / output
? / 2948 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
ReportLocalMaxxing		Muse-Glimmer-30BUD-Q8_K_XLModel ref. ↗Token speed—
663.48 tok/sPrefill · unknown
15.95 tok/sGeneration · peak · unknown
3 × R9700Multi-GPU · split unknown · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Arch Linux
Engine
llama.cpp
Backend
rocm
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
578 / 256 tokens
Context
834
Concurrency
Not reported
Date
2026-08-10
Full record →
View
ReportCustom evaluation		Muse-Glimmer-30BQ4_K_MModel ref. ↗Conflicting evidenceToken speedsynth-poincare-section-pytest——
26.91 tok/sGeneration · mean · per request
1 × W7800Single GPU · 48 GB VRAM/GPU
Configuration
Memory
48 GB VRAM/GPU
OS
Ubuntu 24.04 LTS
Engine
Ollama
Backend
ROCm
Harness
BoJl unified evaluation harness (2026-08)
Prompt / output
? / 3552 tokens
Context
Not reported
Concurrency
Not reported
Date
Not reported
Full record →
View
ReportLocalMaxxing		Muse-Glimmer-30BUD-Q8_K_XLModel ref. ↗Token speed—
660.07 tok/sPrefill · unknown
15.99 tok/sGeneration · peak · unknown
3 × R9700Multi-GPU · split unknown · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Arch Linux
Engine
llama.cpp
Backend
rocm
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
578 / 256 tokens
Context
834
Concurrency
Not reported
Date
2026-08-10
Full record →
Quantizations and software

Quantizations: Dynamic-Q4_K_XL + DFlash2-Q4_K_M, FP8, Q4, Q4_K_M, Q4_K_XL, Q5_K_XL, ROCMFP4_FAST, UD-Q2-K-XL, UD-Q8_K_XL
Engines: llama.cpp, Ollama, vllm

Explore all 45 results →

This is a selection across sources, not a ranking or a matched experiment. Reading benchmark reports

GPU benchmark pages: W7800 · R9700 · RX 7600 XT · Ryzen AI Max+ 395