Benchmark result
unsloth/Phi-4-mini-instruct-GGUF on Intel(R) Arc(TM) 140V GPU (16GB) — 26.6 tok/s
Measured with Unsloth Studio on October 6, 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 Intel(R) Arc(TM) 140V GPU (16GB) · median 26.4 tok/s.
sort
- 126.6tok/s36.8Unsloth Studio · Q4_K_M · KV q8_0this run
- 226.1tok/s45.0Unsloth Studio · Q4_K_M · KV q4_0
Reproduce this run
toolUnsloth Studio
modelunsloth/Phi-4-mini-instruct-GGUF (Q4_K_M)
context32,768
kv cacheq8_0
think limitnone
samplingtemperature 0.8 · top_p 0.95 · top_k -1 · min_p 0 · presence_penalty 0
clientv0.4.89+102
Set those in Unsloth Studio, then:
llm-benchmark benchmark --model "unsloth/Phi-4-mini-instruct-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.