Benchmark result
gemma-4-26B-A4B-it-UD-IQ4_NL on 2× NVIDIA GeForce RTX 5060 — 93.8 tok/s
Measured with llama.cpp b10918-82d6bb284 on September 12, 2026.
What the model built
Open full screen →The coding scenario asks for a playable game in a single HTML file. This is exactly what the model returned, unedited; the frame only adds a line that tells this page how tall it is.
How this run compares
1st fastest of 3 runs of this model on 2× NVIDIA GeForce RTX 5060 · Multi-GPU · median 78.2 tok/s.
sort
- 193.8tok/s74.3llama.cppthis run
- 278.2tok/s75.6llama.cpp
- 377.6tok/s73.7llama.cpp
Reproduce this run
toolllama.cpp b10918-82d6bb284
modelgemma-4-26B-A4B-it-UD-IQ4_NL
context131,072
thinkingon
samplingtemperature 0.20000000298023224 · top_p 0.8999999761581421 · top_k 0 · min_p 0.05000000074505806 · presence_penalty 0 · repeat_penalty 1
clientv0.4.59+97
Set those in llama.cpp, then:
llm-benchmark benchmark --model "gemma-4-26B-A4B-it-UD-IQ4_NL" --tool "llama.cpp"The client prompts for context length and thinking mode, and for the KV cache dtype on Unsloth Studio. Sampling is left at the model default — the values above are what the tool reported using, not overrides the client sent.