Benchmark result
occamy-1.0-abliterated-PXQN3 on 2× NVIDIA GeForce GTX 1070 — 23.9 tok/s
Measured with llama.cpp b10912-e37f292792 on November 19, 2026.
Support llm-bench.io
llm-bench.io is free. If it helped you, I'd appreciate your support to cover the infrastructure costs.
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 GTX 1070 · Multi-GPU · median 18.3 tok/s.
sort
- 123.9tok/s80.8llama.cppthis run
- 219.5tok/s84.0llama.cpp
- 317.1tok/s85.9llama.cpp
- 412.4tok/s82.9llama.cpp
Reproduce this run
toolllama.cpp b10912-e37f292792
modeloccamy-1.0-abliterated-PXQN3
context65,536
thinkingon
think limit4,096 tokens
samplingtemperature 0.699999988079071 · top_p 0.800000011920929 · top_k 20 · min_p 0 · presence_penalty 0 · repeat_penalty 1
clientv0.4.99+102
Set those in llama.cpp, then:
llm-benchmark benchmark --model "occamy-1.0-abliterated-PXQN3" --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.