AMD-Pro-BenchCommunity

13,812 benchmarks in 32 GPU combinations documented

Qwen3.8-Flash-Next-R9V on R9700

293 results · 9 sources · 293 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
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed——
36.03 tok/sGeneration · mean · per request
2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 44
Engine
R9V / vLLM
Backend
Not reported
Harness
Not reported
Prompt / output
278 / 256 tokens
Context
131072
Concurrency
1
Date
2026-09-02
Full record →
View
ReportGeneric LLM		Qwen3.8-Flash-Next-R9VIQ4_XSWeights ↗Token speed—
489.53 tok/sPrefill · mean · per request
—2 × R9700Layer split · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
llama.cpp
Backend
ROCm 7.14
Harness
Not reported
Prompt / output
271 / 256 tokens
Context
Not reported
Concurrency
1
Date
2026-08-30
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed—
1,005.35 tok/sPrefill · per request
—2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
R9V / vLLM
Backend
Not reported
Harness
scripts/bench-pp.py
Prompt / output
8192 / 1 tokens
Context
131072
Concurrency
1
Date
2026-09-14
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed—
1,621.76 tok/sPrefill · per request
—2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
R9V / vLLM
Backend
Not reported
Harness
scripts/bench-pp.py
Prompt / output
8192 / 1 tokens
Context
131072
Concurrency
1
Date
2026-08-30
Full record →
View
ReportGeneric LLM		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed——
26.22 tok/sGeneration · mean · per request
2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
public vLLM Radiance comparator
Backend
Not reported
Harness
Not reported
Prompt / output
278 / 256 tokens
Context
8448
Concurrency
1
Date
2026-08-30
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed—
1,668.2 tok/sPrefill · per request
—2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
R9V / vLLM
Backend
Not reported
Harness
scripts/bench-pp.py
Prompt / output
8192 / 1 tokens
Context
131072
Concurrency
1
Date
2026-09-15
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed——
93.02 tok/sGeneration · per request
2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
R9V / vLLM
Backend
Not reported
Harness
Not reported
Prompt / output
? / ? tokens
Context
131072
Concurrency
1
Date
2026-09-12
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed——
90.22 tok/sGeneration · per request
2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
R9V / vLLM
Backend
Not reported
Harness
Not reported
Prompt / output
? / ? tokens
Context
131072
Concurrency
1
Date
Not reported
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VIQ4_XS target / FP8 MTPWeights ↗Token speed—
2,361 tok/sPrefill · median · per request
—2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Not reported
Engine
R9V / vLLM
Backend
ROCm 10
Harness
Not reported
Prompt / output
8192 / ? tokens
Context
131072
Concurrency
1
Date
Not reported
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed——
35.85 tok/sGeneration · median · per request
2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 44
Engine
R9V / vLLM
Backend
Not reported
Harness
Not reported
Prompt / output
278 / 256 tokens
Context
131072
Concurrency
1
Date
2026-09-02
Full record →
View
ReportGeneric LLM		Qwen3.8-Flash-Next-R9VIQ4_XSWeights ↗Token speed—
461.49 tok/sPrefill · per request
—2 × R9700Layer split · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
llama.cpp
Backend
ROCm 7.14
Harness
Not reported
Prompt / output
271 / 256 tokens
Context
Not reported
Concurrency
1
Date
2026-08-30
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed—
1,014.62 tok/sPrefill · per request
—2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
R9V / vLLM
Backend
Not reported
Harness
scripts/bench-pp.py
Prompt / output
8192 / 1 tokens
Context
131072
Concurrency
1
Date
2026-09-14
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed—
1,477.88 tok/sPrefill · per request
—2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
R9V / vLLM
Backend
Not reported
Harness
scripts/bench-pp.py
Prompt / output
8192 / 1 tokens
Context
131072
Concurrency
1
Date
2026-08-30
Full record →
View
ReportGeneric LLM		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed——
26.2 tok/sGeneration · range lower · per request
2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
public vLLM Radiance comparator
Backend
Not reported
Harness
Not reported
Prompt / output
278 / 256 tokens
Context
8448
Concurrency
1
Date
2026-08-30
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed—
1,718.26 tok/sPrefill · per request
—2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
R9V / vLLM
Backend
Not reported
Harness
scripts/bench-pp.py
Prompt / output
8192 / 1 tokens
Context
131072
Concurrency
1
Date
2026-09-15
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed——
93.29 tok/sGeneration · per request
2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
R9V / vLLM
Backend
Not reported
Harness
Not reported
Prompt / output
? / ? tokens
Context
131072
Concurrency
1
Date
2026-09-12
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed——
89.45 tok/sGeneration · per request
2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
R9V / vLLM
Backend
Not reported
Harness
Not reported
Prompt / output
? / ? tokens
Context
131072
Concurrency
1
Date
Not reported
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VIQ4_XS target / FP8 MTPWeights ↗Token speed——
75.8 tok/sGeneration · median · per request
2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Not reported
Engine
R9V / vLLM
Backend
ROCm 10
Harness
Not reported
Prompt / output
8192 / ? tokens
Context
131072
Concurrency
1
Date
Not reported
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed——
35.59 tok/sGeneration · range lower · per request
2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Fedora 44
Engine
R9V / vLLM
Backend
Not reported
Harness
Not reported
Prompt / output
278 / 256 tokens
Context
131072
Concurrency
1
Date
2026-09-02
Full record →
View
ReportGeneric LLM		Qwen3.8-Flash-Next-R9VIQ4_XSWeights ↗Token speed—
504.69 tok/sPrefill · per request
—2 × R9700Layer split · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
llama.cpp
Backend
ROCm 7.14
Harness
Not reported
Prompt / output
271 / 256 tokens
Context
Not reported
Concurrency
1
Date
2026-08-30
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed—
967.55 tok/sPrefill · per request
—2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
R9V / vLLM
Backend
Not reported
Harness
scripts/bench-pp.py
Prompt / output
8192 / 1 tokens
Context
131072
Concurrency
1
Date
2026-09-14
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed—
1,486.43 tok/sPrefill · per request
—2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
R9V / vLLM
Backend
Not reported
Harness
scripts/bench-pp.py
Prompt / output
8192 / 1 tokens
Context
131072
Concurrency
1
Date
2026-08-30
Full record →
View
ReportGeneric LLM		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed——
26.25 tok/sGeneration · range upper · per request
2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
public vLLM Radiance comparator
Backend
Not reported
Harness
Not reported
Prompt / output
278 / 256 tokens
Context
8448
Concurrency
1
Date
2026-08-30
Full record →
View
ReportServing benchmark		Qwen3.8-Flash-Next-R9VUD-IQ4_XSWeights ↗Token speed—
1,683.8 tok/sPrefill · per request
—2 × R9700Tensor parallel · 32 GB VRAM/GPU
Configuration
Memory
32 GB VRAM/GPU
OS
Linux (inferred; host distribution unreported) (inferred)
Engine
R9V / vLLM
Backend
Not reported
Harness
scripts/bench-pp.py
Prompt / output
8192 / 1 tokens
Context
131072
Concurrency
1
Date
2026-09-15
Full record →
Quantizations and software

Quantizations: IQ4_XS, IQ4_XS target / FP8 MTP, UD-IQ4_XS
Engines: llama.cpp, public vLLM Radiance comparator, R9V / vLLM

Explore all 293 results →

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

GPU benchmark pages: R9700