Grouped by reported model identity. Exact weights, quantizations and software can differ. How results are reviewed
Documented results
4 selected tests| View Reportllm-bench.io | gemma4:26b-a4b-it-qat | llm-bench.io QualityFour sequential scenarios: agent workflow, code generation, role play, research | 70.11 / 100LLM judge · Mean | 567.17 tok/sPrefill · mean · per request | 32.93 tok/sGeneration · mean · per request | GPU count unknownUnknown · 16 GB VRAM/GPUConfiguration
|
|---|---|---|---|---|---|---|
| View Reportllm-bench.io | gemma4:26b-a4b-it-qat | llm-bench.io QualityFour sequential scenarios: agent workflow, code generation, role play, research | 73.44 / 100LLM judge · Mean | — | 106.24 tok/sGeneration · mean · per request | GPU count unknownUnknown · 24 GB VRAM/GPUConfiguration
|
| View Reportllm-bench.io | gemma4:26b-a4b-it-qat | llm-bench.io QualityFour sequential scenarios: agent workflow, code generation, role play, research | 71.76 / 100LLM judge · Mean | — | 120.28 tok/sGeneration · mean · per request | GPU count unknownUnknown · 24 GB VRAM/GPUConfiguration
|
| View Reportllm-bench.io | gemma4:26b-a4b-it-qat | llm-bench.io QualityFour sequential scenarios: agent workflow, code generation, role play, research | 62.28 / 100LLM judge · Mean | — | 121.41 tok/sGeneration · mean · per request | GPU count unknownUnknown · 24 GB VRAM/GPUConfiguration
|
Quantizations and software
Quantizations: Q4_0
Engines: Ollama
This is a selection across sources, not a ranking or a matched experiment. Reading benchmark reports
GPU benchmark pages: RX 7900 XTX · RX 7800 XT