Benchmark result
prism-ml/Ternary-Bonsai-8B-gguf:Q2_0 on NVIDIA GeForce RTX 4070 — 153.3 tok/s
Measured with llama.cpp manual on October 10, 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 NVIDIA GeForce RTX 4070 · median 151.0 tok/s.
sort
- 1153.3tok/s41.9llama.cpp · Q2_0this run
- 2151.0tok/s47.2llama.cpp · Q2_0
- 3149.0tok/s56.3llama.cpp · Q2_0
Reproduce this run
toolllama.cpp manual
modelprism-ml/Ternary-Bonsai-8B-gguf:Q2_0 (Q2_0)
context65,536
thinkingon
think limitnone
clientv0.4.91+102
Set those in llama.cpp, then:
llm-benchmark benchmark --model "prism-ml/Ternary-Bonsai-8B-gguf:Q2_0" --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.