Benchmark result
peculiar-ragdoll/Tiel-Coder-35B-A3B-GGUF-MTP on Advanced Micro Devices, Inc. [AMD/ATI] HawkPoint1 — 26.8 tok/s
Measured with llama.cpp b10636-4d19b2876 on September 1, 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
2nd fastest of 4 runs of this model on Advanced Micro Devices, Inc. [AMD/ATI] HawkPoint1 · median 26.8 tok/s.
sort
- 126.9tok/s79.3llama.cpp
- 226.8tok/s79.9llama.cppthis run
- 326.8tok/s77.0llama.cpp
- 426.5tok/s79.5llama.cpp
peculiar-ragdoll/Tiel-Coder-35B-A3B-GGUF-MTP on other hardware
Reproduce this run
toolllama.cpp b10636-4d19b2876
modelpeculiar-ragdoll/Tiel-Coder-35B-A3B-GGUF-MTP
context65,536
thinkingoff
clientv0.4.46+97
Set those in llama.cpp, then:
llm-benchmark benchmark --model "peculiar-ragdoll/Tiel-Coder-35B-A3B-GGUF-MTP" --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.