Benchmark result
qwen3.6-moe-35b-a3b-FLM:latest on AMD Radeon 880M / 890M — 17.1 tok/s
Measured with Ollama 0.16.1 on August 28, 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.
Reproduce this run
toolOllama 0.16.1
modelqwen3.6-moe-35b-a3b-FLM:latest
context32,768
clientv0.4.44+97
Set those in Ollama, then:
llm-benchmark benchmark --model "qwen3.6-moe-35b-a3b-FLM:latest" --tool "Ollama"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.