Benchmark result
Ternary-Bonsai-2-27B-PQ2_0-MTP-Q8_0 on AMD Radeon Pro W7900 — 61.3 tok/s
Measured with llama.cpp b10697-f0a2b5dc9 on September 19, 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 2 runs of this model on AMD Radeon Pro W7900 · median 60.6 tok/s.
sort
- 161.3tok/s85.4llama.cpp · Q2_0this run
- 259.9tok/s81.9llama.cpp · Q2_0
Reproduce this run
toolllama.cpp b10697-f0a2b5dc9
modelTernary-Bonsai-2-27B-PQ2_0-MTP-Q8_0 (Q2_0)
context65,536
thinkingon
effortxhigh (model default)
samplingtemperature 1 · top_p 0.949999988079071 · top_k 20 · min_p 0 · presence_penalty 0 · repeat_penalty 1
clientv0.4.59+97
Set those in llama.cpp, then:
llm-benchmark benchmark --model "Ternary-Bonsai-2-27B-PQ2_0-MTP-Q8_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.