Benchmark result
protoLabsAI/Ornith-1.5-9B-MTP-GGUF:Q5_K_M on AMD Radeon RX 9070/9070 XT/9070 GRE — 92.7 tok/s
Measured with llama.cpp b11388-050439614 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 5 runs of this model on AMD Radeon RX 9070/9070 XT/9070 GRE · median 79.1 tok/s.
sort
- 192.7tok/s62.8llama.cpp · Q5_K_Mthis run
- 282.6tok/s75.9llama.cpp · Q5_K_M
- 379.1tok/s79.2llama.cpp · Q5_K_M
- 474.5tok/s77.2llama.cpp · Q5_K_M
- 565.7tok/s77.7llama.cpp · Q5_K_M
Reproduce this run
toolllama.cpp b11388-050439614
modelprotoLabsAI/Ornith-1.5-9B-MTP-GGUF:Q5_K_M (Q5_K_M)
context32,768
think limitnone
samplingtemperature 1 · top_p 0.949999988079071 · top_k 20 · min_p 0.05000000074505806 · presence_penalty 0 · repeat_penalty 1
clientv0.4.87+102
Set those in llama.cpp, then:
llm-benchmark benchmark --model "protoLabsAI/Ornith-1.5-9B-MTP-GGUF:Q5_K_M" --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.