RX 9060 XT LLM benchmarks
73 results · 103 measurements · 11 sources
Device specifications
| Variant | Memory | Memory type | Bus | Bandwidth | FP16 matrix | INT8 matrix |
|---|---|---|---|---|---|---|
| RX 9060 XT 16GB ↗ | 16 GB | GDDR6 | 128-bit | 320 GB/s | 103 TFLOPS | 205 TOPS |
Manufacturer specifications. Shared system RAM is not dedicated VRAM. Unreported values remain —; measured LLM results retain their reported configuration.
Selected reported tests
24 of 73 resultsSource-reported results, selected across reports by documentation quality. Match configurations before comparing values.
| View Reportllm-bench.io | Qwen3.8-27B | llm-bench.io QualityFour sequential scenarios: agent workflow, code generation, role play, research | 81.27 / 100LLM judge · Mean | 171.11 tok/sPrefill · mean · per request | 5.78 tok/sGeneration · mean · per request | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
|---|---|---|---|---|---|---|
| View ReportLocalMaxxing | DeepSeek-R1-Distill-Qwen-7B | Token speed | — | 1,844.13 tok/sPrefill · unknown | 39.76 tok/sGeneration · unknown | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Qwen2.5-Coder-14B-Instruct | Token speed | — | 1,232.42 tok/sPrefill · mean · per request | — | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Qwen2.5-0.5B-Instruct (identity conflict)Conflicting evidence | Token speed | — | 7,898.02 tok/sPrefill · mean · per request | — | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Gemma 4 12B | Token speed | — | 1,467 tok/sPrefill · per request | — | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Llama-2-7B | Token speed | — | 2,844.8 tok/sPrefill · mean · per request | — | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Llama-3.1-8B-Instruct | Token speed | — | — | 58.03 tok/sGeneration · per request | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | Llama-2-13b-chat | Procyon | — | — | 34.65 tok/sGeneration · per request | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | MLPerf Client v1.0 (selected models unreported) | Token speed1.0 · MLPerf Client | — | — | 73 tok/sGeneration · per request | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View ReportServing benchmark | Qwen3.8-27BConflicting evidence | Other LLM measurements1 concurrent request | 15.2 updates/sStream update rate · Per request | — | — | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | Llama-3.1-8B-Instruct | Procyon | 1,213 pointsBenchmark score · Per request | — | — | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View ReportLocalMaxxing | DeepSeek-R1-Distill-Qwen-7B | Token speed | — | 2,312.75 tok/sPrefill · unknown | 38.13 tok/sGeneration · unknown | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Qwen2.5-Coder-14B-Instruct | Token speed | — | — | 31.33 tok/sGeneration · mean · per request | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Qwen2.5-0.5B-Instruct (identity conflict)Conflicting evidence | Token speed | — | — | 200.71 tok/sGeneration · mean · per request | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Gemma 4 12B | Token speed | — | — | 37.6 tok/sGeneration · per request | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Llama-2-7B | Token speed | — | — | 71.07 tok/sGeneration · mean · per request | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Llama-3.1-8B-Instruct | Token speed | — | — | 75.08 tok/sGeneration · per request | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | Llama-3.1-8B-Instruct | Procyon | — | — | 53.17 tok/sGeneration · per request | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | MLPerf Client v1.0 (selected models unreported) | Latency1.0 · MLPerf Client | 680 msTime to first token · Per request | — | — | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View ReportServing benchmark | Qwen3.8-27BConflicting evidence | Other LLM measurements1 concurrent request | 28.7 updates/sStream update rate · Per request | — | — | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View ReportLocalMaxxing | gemma-4-E4B | Token speed | — | 2,118.5 tok/sPrefill · unknown | 52.73 tok/sGeneration · unknown | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Qwen2.5-Coder-14B-Instruct | Token speed | — | 1,257.44 tok/sPrefill · mean · per request | — | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Qwen2.5-0.5B-Instruct (identity conflict)Conflicting evidence | Token speed | — | 8,211.17 tok/sPrefill · mean · per request | — | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Gemma 4 12B | Token speed | — | 1,509 tok/sPrefill · per request | — | 1 × RX 9060 XTSingle GPU · 16 GB VRAM/GPUConfiguration
|
Models and software in this collection
Model families: DeepSeek R1, GPT-OSS, Gemma 4, Llama 2, Llama 3.1, Mistral, Phi 3.5, Qwen2.5, Qwen3.5, Qwen3.8
Engines: LM Studio, ONNX Runtime, llama.cpp
Methodology and evidence standards · Download the research dataset