Benchmark result
tiel-coder-35b-a3b-mtp-ud-iq4_xs on NVIDIA GPU +1 more — 87.2 tok/s
Measured with llama.cpp PolyStrata 0.1.45 on November 15, 2026.
Support llm-bench.io
llm-bench.io is free. If it helped you, I'd appreciate your support to cover the infrastructure costs.
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 NVIDIA GPU +1 more · Multi-GPU · median 86.8 tok/s.
sort
- 187.2tok/s87.0llama.cppthis run
- 286.3tok/s83.1llama.cpp
Reproduce this run
toolllama.cpp PolyStrata 0.1.45
modeltiel-coder-35b-a3b-mtp-ud-iq4_xs
context131,072
thinkingon
clientv0.4.99+102
Set those in llama.cpp, then:
llm-benchmark benchmark --model "tiel-coder-35b-a3b-mtp-ud-iq4_xs" --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.