AMD-Pro-BenchCommunity

13,812 benchmarks in 32 GPU combinations documented

3 × W7900 LLM benchmarks

5 results · 5 measurements · 1 source

Device specifications
VariantMemoryMemory typeBusBandwidthFP16 matrixINT8 matrix
W7900 Dual Slot ↗48 GBGDDR6384-bit864 GB/s123 TFLOPS123 TOPS
W7900 ↗48 GBGDDR6384-bit864 GB/s123 TFLOPS123 TOPS

Manufacturer specifications. Shared system RAM is not dedicated VRAM. Unreported values remain —; measured LLM results retain their reported configuration.

Selected reported tests

5 of 5 results

Source-reported results, selected across reports by documentation quality. Match configurations before comparing values.

View
Report
Generic LLM
HunyuanOCRBF16Weights ↗OmniDocBenchFull1651 pages · v1.6
94.05 %Overall · n = 100 · Official page-level aggregates
——3 × W7900Independent replicas · 48 GB VRAM/GPU
Configuration
Memory
48 GB VRAM/GPU
OS
Linux
Engine
Transformers
Backend
ROCm / HIP
Harness
OmniDocBench quick_match + CDM
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-07-31
Full record →
View
Report
Generic LLM
HunyuanOCRBF16Weights ↗OmniDocBenchFull1651 pages · v1.6
0.04 edit distanceText edit distance · Official page-level aggregates · Lower is better
——3 × W7900Independent replicas · 48 GB VRAM/GPU
Configuration
Memory
48 GB VRAM/GPU
OS
Linux
Engine
Transformers
Backend
ROCm / HIP
Harness
OmniDocBench quick_match + CDM
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-07-31
Full record →
View
Report
Generic LLM
HunyuanOCRBF16Weights ↗OmniDocBenchFull1651 pages · v1.6
93.33 %Table TEDS · n = 100 · Official page-level aggregates
——3 × W7900Independent replicas · 48 GB VRAM/GPU
Configuration
Memory
48 GB VRAM/GPU
OS
Linux
Engine
Transformers
Backend
ROCm / HIP
Harness
OmniDocBench quick_match + CDM
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-07-31
Full record →
View
Report
Generic LLM
HunyuanOCRBF16Weights ↗OmniDocBenchFull1651 pages · v1.6
92.6 %Formula CDM · n = 100 · Official page-level aggregates
——3 × W7900Independent replicas · 48 GB VRAM/GPU
Configuration
Memory
48 GB VRAM/GPU
OS
Linux
Engine
Transformers
Backend
ROCm / HIP
Harness
OmniDocBench quick_match + CDM
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-07-31
Full record →
View
Report
Generic LLM
HunyuanOCRBF16Weights ↗OmniDocBenchFull1651 pages · v1.6
0.13 edit distanceReading-order edit distance · Official page-level aggregates · Lower is better
——3 × W7900Independent replicas · 48 GB VRAM/GPU
Configuration
Memory
48 GB VRAM/GPU
OS
Linux
Engine
Transformers
Backend
ROCm / HIP
Harness
OmniDocBench quick_match + CDM
Prompt / output
? / ? tokens
Context
Not reported
Concurrency
Not reported
Date
2026-07-31
Full record →
Models and software in this collection

Model families:

Engines: Transformers

Methodology and evidence standards · Download the research dataset