Benchmark result
Cyber-Tiel-Coder-35B-A3B-MTP-UD-IQ3_XXS on 2× NVIDIA GeForce RTX 5060 — 119.3 tok/s
Measured with llama.cpp b10883-91f6a6cf3 on September 9, 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 4 runs of this model on 2× NVIDIA GeForce RTX 5060 · Multi-GPU · median 115.8 tok/s.
sort
- 1119.3tok/s84.0llama.cppthis run
- 2117.4tok/s70.4llama.cpp
- 3114.2tok/s78.3llama.cpp
- 4113.5tok/s82.2llama.cpp
Cyber-Tiel-Coder-35B-A3B-MTP-UD-IQ3_XXS on other hardware
Reproduce this run
toolllama.cpp b10883-91f6a6cf3
modelCyber-Tiel-Coder-35B-A3B-MTP-UD-IQ3_XXS
context131,072
thinkingon
samplingtemperature 0.6000000238418579 · top_p 0.8999999761581421 · top_k 0 · min_p 0.05000000074505806 · presence_penalty 0 · repeat_penalty 1
clientv0.4.58+97
Set those in llama.cpp, then:
llm-benchmark benchmark --model "Cyber-Tiel-Coder-35B-A3B-MTP-UD-IQ3_XXS" --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.