RX 7900 GRE LLM benchmarks
56 results · 98 measurements · 6 sources
Device specifications
| Variant | Memory | Memory type | Bus | Bandwidth | FP16 matrix | INT8 matrix |
|---|---|---|---|---|---|---|
| RX 7900 GRE ↗ | 16 GB | GDDR6 | 256-bit | 576 GB/s | 92 TFLOPS | 92 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 56 resultsSource-reported results, selected across reports by documentation quality. Match configurations before comparing values.
| View Reportllm-bench.io | Qwen3.5-9B | llm-bench.io QualityFour sequential scenarios: agent workflow, code generation, role play, research | 62.5 / 100LLM judge · Mean | 1,029.22 tok/sPrefill · mean · per request | 70.18 tok/sGeneration · mean · per request | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
|---|---|---|---|---|---|---|
| View ReportCoding session | Qwen3.8-27B | Token speed | — | 246.8 tok/sPrefill · per request | 34.08 tok/sGeneration · per request | 1 × RX 7900 GRESingle GPU · 16 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | Moonlight-16B-A3B-Instruct | Token speed | — | 188.1 tok/sPrefill · per request | 100.2 tok/sGeneration · per request | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | DeepSeek-R1-Distill-Llama-8B | Token speed | — | 2,234.19 tok/sPrefill · mean · per request | — | 1 × RX 7900 GRESingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Llama-2-7B | Token speed | — | 1,456.98 tok/sPrefill · mean · per request | — | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Llama-2-7B | Token speed | — | 2,336.31 tok/sPrefill · mean · per request | — | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
| View Reportllm-bench.io | qwen3.8-27b-gsq-rco@iq2_xs | llm-bench.io QualityFour sequential scenarios: agent workflow, code generation, role play, research | 87.25 / 100LLM judge · Mean | 363.3 tok/sPrefill · mean · per request | 27.97 tok/sGeneration · mean · per request | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
| View ReportCoding session | Qwen3.8-27B | Token speed | — | 460.7 tok/sPrefill · per request | 30.94 tok/sGeneration · per request | 1 × RX 7900 GRESingle GPU · 16 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | gemma-4-26B-A4B-it | Token speed | — | 80.7 tok/sPrefill · per request | 17.3 tok/sGeneration · per request | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | DeepSeek-R1-Distill-Llama-8B | Token speed | — | — | 60.68 tok/sGeneration · mean · per request | 1 × RX 7900 GRESingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Llama-2-7B | Token speed | — | — | 96.07 tok/sGeneration · mean · per request | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Llama-2-7B | Token speed | — | — | 116.11 tok/sGeneration · mean · per request | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
| View Reportllm-bench.io | qwen3.8-27b-gsq-rco@iq2_xs | llm-bench.io QualityFour sequential scenarios: agent workflow, code generation, role play, research | 73.43 / 100LLM judge · Mean | 448.83 tok/sPrefill · mean · per request | 33.89 tok/sGeneration · mean · per request | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
| View ReportCoding session | Qwen3.8-27B | Token speed | — | 370.8 tok/sPrefill · per request | 30.58 tok/sGeneration · per request | 1 × RX 7900 GRESingle GPU · 16 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | gemma-4-26B-A4B-it | Token speed | — | 61.3 tok/sPrefill · per request | 31 tok/sGeneration · per request | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | gpt-oss-20b | Token speed | — | 2,818.34 tok/sPrefill · mean · per request | — | 1 × RX 7900 GRESingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Llama-2-7B | Token speed | — | 1,598.79 tok/sPrefill · mean · per request | — | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Llama-2-7B | Token speed | — | 2,251.37 tok/sPrefill · mean · per request | — | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
| View Reportllm-bench.io | qwen3.8-27b-gsq-rco@iq2_xs | llm-bench.io QualityFour sequential scenarios: agent workflow, code generation, role play, research | 66.19 / 100LLM judge · Mean | 450.9 tok/sPrefill · mean · per request | 31.87 tok/sGeneration · mean · per request | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
| View ReportCoding session | Qwen3.8-27B | Token speed | — | 261.7 tok/sPrefill · per request | 30.94 tok/sGeneration · per request | 1 × RX 7900 GRESingle GPU · 16 GB VRAM/GPUConfiguration
|
| View ReportGeneric LLM | Qwen3.6-35B-A3B | Token speed | — | 41.7 tok/sPrefill · per request | 22.7 tok/sGeneration · per request | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | gpt-oss-20b | Token speed | — | — | 151.4 tok/sGeneration · mean · per request | 1 × RX 7900 GRESingle GPU · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Llama-2-7B | Token speed | — | — | 97.53 tok/sGeneration · mean · per request | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
| View Reportllama-bench | Llama-2-7B | Token speed | — | — | 110.5 tok/sGeneration · mean · per request | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
Models and software in this collection
Model families: DeepSeek R1, GPT-OSS, Gemma 4, Granite 3.0, Llama 2, Llama 3.1, Mistral, Qwen3.5, Qwen3.6, Qwen3.8
Engines: LM Studio, Unsloth Studio, llama.cpp
Methodology and evidence standards · Download the research dataset