Benchmark result

prism-ml/Ternary-Bonsai-2-4B-gguf on NVIDIA GeForce RTX 4070 — 155.5 tok/s

Measured with Unsloth Studio manual on October 16, 2026.

What the model built

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

prism-ml/Ternary-Bonsai-2-4B-gguf on other hardware

38.5tok/s54.5Intel(R) Arc(TM) 140V GPU (16GB) · Q2_0 · KV q8_0

Reproduce this run

toolUnsloth Studio manual

modelprism-ml/Ternary-Bonsai-2-4B-gguf (Q2_0)

context65,536

kv cacheq8_0

thinkingon

think limitnone

samplingtemperature 0.7 · top_p 0.95 · top_k -1 · min_p 0.01 · presence_penalty 0

clientv0.4.96+102

Set those in Unsloth Studio, then:

llm-benchmark benchmark --model "prism-ml/Ternary-Bonsai-2-4B-gguf" --tool "Unsloth Studio"

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.