AMD-Pro-BenchCommunity

13,812 benchmarks in 32 GPU combinations documented

Qwen3.5-4B on AMD GPUs

22 results · 3 sources · 39 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
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · code-debug
120.6 tok/sThroughput · mean · unknown
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
Vulkan / Mesa RADV
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportLocalMaxxing		Qwen3.5-4BIQ4_NLModel ref. ↗Token speed——
61 tok/sGeneration · unknown
1 × Ryzen AI Max+ 395Single GPU · Unified memory
Configuration
Memory
Unified memory
OS
Linux
Engine
llama.cpp
Backend
rocm
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
53 / 1024 tokens
Context
131072
Concurrency
Not reported
Date
2026-05-01
Full record →
View
ReportLocalMaxxing		Qwen3.5-4BQ8_0Model ref. ↗Token speed——
84.86 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
vulkan
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
531 / 256 tokens
Context
787
Concurrency
Not reported
Date
2026-05-07
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · code-debug
106.7 tok/sThroughput · mean · unknown
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
AMD ROCm
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportLocalMaxxing		Qwen3.5-4BIQ4_NLModel ref. ↗Token speed——
63.92 tok/sGeneration · unknown
1 × Ryzen AI Max+ 395Single GPU · Unified memory
Configuration
Memory
Unified memory
OS
Linux
Engine
llama.cpp
Backend
vulkan
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
53 / 1024 tokens
Context
131072
Concurrency
Not reported
Date
2026-05-01
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · code-short
113.4 tok/sThroughput · mean · unknown
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
Vulkan / Mesa RADV
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportLocalMaxxing		Qwen3.5-4BIQ4_NLModel ref. ↗Token speed——
63.89 tok/sGeneration · unknown
1 × Ryzen AI Max+ 395Single GPU · Unified memory
Configuration
Memory
Unified memory
OS
Linux
Engine
llama.cpp
Backend
vulkan
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
3663 / 2048 tokens
Context
262144
Concurrency
Not reported
Date
2026-04-29
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · code-short
105.5 tok/sThroughput · mean · unknown
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
AMD ROCm
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportLocalMaxxing		Qwen3.5-4BUD-Q4_K_XLModel ref. ↗Token speed——
58.31 tok/sGeneration · unknown
1 × Ryzen AI Max+ 395Single GPU · Unified memory
Configuration
Memory
Unified memory
OS
Linux
Engine
llama.cpp
Backend
vulkan
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
3663 / 2048 tokens
Context
262144
Concurrency
Not reported
Date
2026-04-29
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · code-explain
122.1 tok/sThroughput · mean · unknown
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
Vulkan / Mesa RADV
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportLocalMaxxing		Qwen3.5-4BIQ4_NLModel ref. ↗Token speed——
57.12 tok/sGeneration · unknown
1 × Ryzen AI Max+ 395Single GPU · Unified memory
Configuration
Memory
Unified memory
OS
Linux
Engine
llama.cpp
Backend
rocm
Harness
LocalMaxxing community speed test; exact client where reported
Prompt / output
3663 / 2048 tokens
Context
262144
Concurrency
Not reported
Date
2026-04-29
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · code-explain
106.2 tok/sThroughput · mean · unknown
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
AMD ROCm
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · chat-long-output
125.8 tok/sThroughput · mean · unknown
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
Vulkan / Mesa RADV
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · chat-long-output
108.2 tok/sThroughput · mean · unknown
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
AMD ROCm
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · code-debug
195.4 msTime to first token · Per request
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
Vulkan / Mesa RADV
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · code-debug
452.8 msTime to first token · Per request
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
AMD ROCm
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · code-short
86.2 msTime to first token · Per request
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
Vulkan / Mesa RADV
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · code-short
388.9 msTime to first token · Per request
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
AMD ROCm
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · code-explain
124.7 msTime to first token · Per request
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
Vulkan / Mesa RADV
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · code-explain
415.9 msTime to first token · Per request
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
AMD ROCm
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · chat-long-output
87.4 msTime to first token · Per request
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
Vulkan / Mesa RADV
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
View
ReportGeneric LLM		Qwen3.5-4BQuant not reportedWeights ↗Conflicting evidenceLemonade benchmark2026.39.1 · chat-long-output
395.6 msTime to first token · Per request
——1 × R9700Single GPU · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Ubuntu 26.04
Engine
Lemonade / llama.cpp
Backend
AMD ROCm
Harness
Lemonade
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-10-05
Full record →
Quantizations and software

Quantizations: IQ4_NL, Q8_0, UD-Q4_K_XL
Engines: Lemonade / llama.cpp, 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