AMD-Pro-BenchCommunity

13,812 benchmarks in 32 GPU combinations documented

Qwen3-30B-A3B on AMD GPUs

22 results · 2 sources · 30 measurements

Explore all results

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

Documented results

22 selected tests
View
ReportLocalMaxxing		Qwen3-30B-A3BQ4_K_XLModel ref. ↗Token speed—
1,450.3 tok/sPrefill · median · unknown
89.5 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
1177 / 512 tokens
Context
8192
Concurrency
Not reported
Date
2026-08-12
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Token speed—
1,437.18 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
128 / 0 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
ReportLocalMaxxing		Qwen3-30B-A3BQ4_K_XLModel ref. ↗Token speed—
1,416.6 tok/sPrefill · median · unknown
88.5 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
1177 / 512 tokens
Context
8192
Concurrency
Not reported
Date
2026-08-12
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Token speed—
2,196.54 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
256 / 0 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Token speed—
3,007.39 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
512 / 0 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Token speed—
2,914.3 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
1024 / 0 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Conflicting evidenceToken speed——
177.87 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
0 / 128 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Conflicting evidenceToken speed——
178.4 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
0 / 256 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Conflicting evidenceToken speed——
174.82 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
0 / 512 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Conflicting evidenceToken speed——
169.53 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
0 / 1024 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Conflicting evidenceToken speed——
165.27 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
0 / 2048 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Conflicting evidenceToken speed——
156.8 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
0 / 4096 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Token speed—
1,451.97 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
128 / 0 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Token speed—
2,197.18 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
256 / 0 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Token speed—
3,032.59 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
512 / 0 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Token speed—
3,009.02 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
1024 / 0 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Conflicting evidenceToken speed——
183.47 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
0 / 128 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Conflicting evidenceToken speed——
183.76 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
0 / 256 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Conflicting evidenceToken speed——
183.21 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
0 / 512 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Conflicting evidenceToken speed——
181.32 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
0 / 1024 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Conflicting evidenceToken speed——
177.05 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
0 / 2048 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
View
Reportllama-bench		Qwen3-30B-A3BQ4_K_MModel ref. ↗Conflicting evidenceToken speed——
171.3 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4 LTS / Linux 6.17.0-14-generic
Engine
llama.cpp
Backend
Vulkan
Harness
llama-bench
Prompt / output
0 / 4096 tokens
Context
Not reported
Concurrency
1
Date
2026-03-14
Full record →
Quantizations and software

Quantizations: Q4_K_M, Q4_K_XL
Engines: llama.cpp

Explore all 22 results →

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

GPU benchmark pages: R9700 · Ryzen AI Max+ 395