Benchmark result
LFM2.5-24B-A2B.APEX-I-MiniPlus-V2.1 on Intel Graphics — 29.6 tok/s
Measured with llama.cpp manual on October 15, 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 Intel Graphics · median 28.4 tok/s.
sort
- 129.6tok/s70.4llama.cppthis run
- 228.4tok/s56.6llama.cpp:vulkan
- 316.3tok/s68.2llama.cpp:sycl
Reproduce this run
toolllama.cpp manual
modelLFM2.5-24B-A2B.APEX-I-MiniPlus-V2.1
context65,536
thinkingon
think limitnone
samplingtemperature 0.10000000149011612 · top_p 0.949999988079071 · top_k 20 · min_p 0 · presence_penalty 0 · repeat_penalty 1.0499999523162842
clientv0.4.87+102
Set those in llama.cpp, then:
llm-benchmark benchmark --model "LFM2.5-24B-A2B.APEX-I-MiniPlus-V2.1" --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.