Benchmark result
Ternary-Bonsai-2-27B-PTQ1_0 on NVIDIA GeForce RTX 4070 — 57.5 tok/s
Measured with OpenAI-compatible manual on October 8, 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 56.3 tok/s.
sort
- 157.5tok/s78.9OpenAI-compatible · Q1_0this run
- 256.3tok/s77.6llama.cpp · Q1_0
- 328.6tok/s83.2llama.cpp · Q1_0
Ternary-Bonsai-2-27B-PTQ1_0 on other hardware
Reproduce this run
toolOpenAI-compatible manual
modelTernary-Bonsai-2-27B-PTQ1_0 (Q1_0)
context65,536
thinkingon
clientv0.4.89+102
Set those in OpenAI-compatible, then:
llm-benchmark benchmark --model "Ternary-Bonsai-2-27B-PTQ1_0" --tool "OpenAI-compatible"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.