AMD-Pro-BenchCommunity

13,812 benchmarks in 32 GPU combinations documented

Qwen2.5-14B-Instruct on AMD GPUs

42 results · 4 sources · 96 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
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
591.81 tok/sPrefill · mean · unknown
37.7 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 24 GB VRAM/GPU
Configuration
Memory
24 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
1024 / 16 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-04-19
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
592.73 tok/sPrefill · mean · unknown
4.16 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 24 GB VRAM/GPU
Configuration
Memory
24 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
1024 / 16 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-06-26
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
493.61 tok/sPrefill · mean · unknown
4.58 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
1024 / 16 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-07-04
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
1,386 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4
Engine
llama.cpp
Backend
HIP / ROCm 7.2.0
Harness
Not reported
Prompt / output
128 / ? tokens
Context
8192
Concurrency
Not reported
Date
Not reported
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
461.97 tok/sPrefill · mean · unknown
19.27 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 24 GB VRAM/GPU
Configuration
Memory
24 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
4096 / 256 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-04-19
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
448.25 tok/sPrefill · mean · unknown
1.15 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 24 GB VRAM/GPU
Configuration
Memory
24 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
4096 / 256 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-06-26
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
407.12 tok/sPrefill · mean · unknown
1.15 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
4096 / 256 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-07-04
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
1,675 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4
Engine
llama.cpp
Backend
HIP / ROCm 7.2.0
Harness
Not reported
Prompt / output
256 / ? tokens
Context
8192
Concurrency
Not reported
Date
Not reported
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
537.22 tok/sPrefill · mean · unknown
28.3 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 24 GB VRAM/GPU
Configuration
Memory
24 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
2048 / 256 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-04-19
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
520.87 tok/sPrefill · mean · unknown
2.96 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 24 GB VRAM/GPU
Configuration
Memory
24 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
2048 / 256 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-06-26
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
446.51 tok/sPrefill · mean · unknown
3.24 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
2048 / 256 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-07-04
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
2,007 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4
Engine
llama.cpp
Backend
HIP / ROCm 7.2.0
Harness
Not reported
Prompt / output
512 / ? tokens
Context
8192
Concurrency
Not reported
Date
Not reported
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
535.98 tok/sPrefill · mean · unknown
26.69 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 24 GB VRAM/GPU
Configuration
Memory
24 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
2048 / 768 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-04-19
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
521.06 tok/sPrefill · mean · unknown
2.7 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 24 GB VRAM/GPU
Configuration
Memory
24 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
2048 / 768 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-06-26
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
443.82 tok/sPrefill · mean · unknown
2.98 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
2048 / 768 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-07-04
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed——
55 tok/sGeneration · mean · per request
1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4
Engine
llama.cpp
Backend
HIP / ROCm 7.2.0
Harness
Not reported
Prompt / output
? / 128 tokens
Context
8192
Concurrency
Not reported
Date
Not reported
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
580.45 tok/sPrefill · mean · unknown
32.96 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 24 GB VRAM/GPU
Configuration
Memory
24 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
1024 / 1024 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-04-19
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
577.84 tok/sPrefill · mean · unknown
3.88 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 24 GB VRAM/GPU
Configuration
Memory
24 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
1024 / 1024 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-06-26
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
467.45 tok/sPrefill · mean · unknown
4.24 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
1024 / 1024 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-07-04
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ5_K_MModel ref. ↗Token speed—
1,322 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4
Engine
llama.cpp
Backend
HIP / ROCm 7.2.0
Harness
Not reported
Prompt / output
128 / ? tokens
Context
8192
Concurrency
Not reported
Date
Not reported
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
557.77 tok/sPrefill · mean · unknown
24.89 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 24 GB VRAM/GPU
Configuration
Memory
24 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
1280 / 3072 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-04-19
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
525.53 tok/sPrefill · mean · unknown
2.23 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 24 GB VRAM/GPU
Configuration
Memory
24 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
1280 / 3072 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-06-26
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ4_K_MModel ref. ↗Token speed—
444.71 tok/sPrefill · mean · unknown
2.45 tok/sGeneration · mean · unknown
GPU count unknownUnknown · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Windows
Engine
llamafile
Backend
Not reported
Harness
LocalScore / llamafile
Prompt / output
1280 / 3072 tokens
Context
Not reported
Concurrency
Not reported
Date
2026-07-04
Full record →
View
ReportGeneric LLM		Qwen2.5-14B-InstructQ5_K_MModel ref. ↗Token speed—
1,383 tok/sPrefill · mean · per request
—1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 24.04.4
Engine
llama.cpp
Backend
HIP / ROCm 7.2.0
Harness
Not reported
Prompt / output
256 / ? tokens
Context
8192
Concurrency
Not reported
Date
Not reported
Full record →
Quantizations and software

Quantizations: Q4_K_M, Q5_K_M, Q8_0
Engines: llama.cpp, llamafile

Explore all 42 results →

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

GPU benchmark pages: R9700 · RX 7900 XTX