Benchmark result
Hermes3.6-35B-A3B-Uncensored-Genesis-V7-APEX on NVIDIA GeForce RTX 3060 +2 more — 57.6 tok/s
Measured with vLLM on October 22, 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.
Reproduce this run
toolvLLM
modelHermes3.6-35B-A3B-Uncensored-Genesis-V7-APEX
context131,072
thinkingon
clientv0.4.97+102
Set those in vLLM, then:
llm-benchmark benchmark --model "Hermes3.6-35B-A3B-Uncensored-Genesis-V7-APEX" --tool "vLLM"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.