Benchmark result

Cyber-Tiel on 2× Tesla V100-PCIE-32GB — 106.5 tok/s

Measured with llama.cpp manual on November 3, 2026.

What the model built

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 2 runs of this model on 2× Tesla V100-PCIE-32GB · Multi-GPU · median 91.0 tok/s.

sort

1106.5tok/s86.4llama.cppthis run
275.5tok/s82.2llama.cpp

Compare the nearest 5 →

Cyber-Tiel on other hardware

Reproduce this run

toolllama.cpp manual

modelCyber-Tiel

context32,768

think limit8,096 tokens

samplingtemperature 0.20000000298023224 · top_p 0.8500000238418579 · top_k 40 · min_p 0.05000000074505806 · presence_penalty 0 · repeat_penalty 1

clientv0.4.98+102

Set those in llama.cpp, then:

llm-benchmark benchmark --model "Cyber-Tiel" --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.