Benchmark result
Huihui-Qwen3.6-35B-A3B-abliterated-MTP-GGUF on NVIDIA GeForce RTX 3070 Ti — 55.1 tok/s
Measured with Unsloth Studio on October 14, 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
10th fastest of 11 runs of this model on NVIDIA GeForce RTX 3070 Ti · median 58.7 tok/s.
sort
- 167.6tok/s80.5Unsloth Studio · Q3_K · KV f16
- 259.2tok/s64.6Unsloth Studio · Q4_K · KV f16
- 359.2tok/s60.4Unsloth Studio · Q4_K · KV f16
- 459.0tok/s66.3Unsloth Studio · Q4_K · KV f16
- 558.9tok/s68.2Unsloth Studio · Q4_K · KV f16
- 658.7tok/s80.4Unsloth Studio · Q4_K · KV f16
- 758.1tok/s64.9Unsloth Studio · Q4_K · KV f16
- 857.5tok/s69.3Unsloth Studio · Q4_K · KV f16
- 956.7tok/s77.6Unsloth Studio · Q4_K · KV f16
- 1055.1tok/s72.2Unsloth Studio · Q4_K · KV f16this run
- 1150.7tok/s77.9Unsloth Studio · Q6_K · KV f16
Reproduce this run
toolUnsloth Studio
modelHuihui-Qwen3.6-35B-A3B-abliterated-MTP-GGUF (Q4_K)
context8,192
kv cachef16 (assumed default)
think limitnone
samplingtemperature 0.7 · top_p 0.8 · top_k 20 · min_p 0 · presence_penalty 1.5
clientv0.4.93+102
Set those in Unsloth Studio, then:
llm-benchmark benchmark --model "Huihui-Qwen3.6-35B-A3B-abliterated-MTP-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.