Model benchmarks

Huihui-Qwen3.6-35B-A3B-abliterated-ggml-model-PXQN3 local LLM performance

As of November 2026, Huihui-Qwen3.6-35B-A3B-abliterated-ggml-model-PXQN3 runs at up to 20.9 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).

llama.cpp
ShareRedditX

Model size

35B

Peak speed

20.9 tok/s

Average speed

19.4 tok/s

Avg PP

328.8 tok/s

Min memory

n/a

Max context

65,536 tokens

Avg output / run

20,592 tokens

Avg runtime / run

18m 30s

Avg quality

77.1

Benchmark runs

2

GPUs tested

1

Quality by task

Average LLM-judged quality (0–100) with the run-to-run spread shown as a P5–P95 band, overall and for each benchmark task, across all 2 runs. The low and high columns show how much the judge’s score varies between runs, and need at least two runs to display.

TaskP5 (low)AvgP95 (high)
Overall74.077.180.3
Agent Workflow73.479.485.4
Code Generation67.171.375.5
Role Play & Narrative87.687.687.6
Research & Analysis67.970.272.4

Performance by hardware and tool

Every hardware/tool/quantization combination Huihui-Qwen3.6-35B-A3B-abliterated-ggml-model-PXQN3 has been benchmarked on, ranked by peak token generation speed. Last updated November 2026.

HardwareToolQuantBest tok/sAvg tok/sMemoryContextQualityRuns
NVIDIA GeForce GTX 1070llama.cpp—20.9 tok/s19.4 tok/sn/a65,536 tokens77.12

Benchmark runs

All 2 Huihui-Qwen3.6-35B-A3B-abliterated-ggml-model-PXQN3 runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.

Frequently asked questions

Is Huihui-Qwen3.6-35B-A3B-abliterated-ggml-model-PXQN3 good for coding?
In our benchmarks, Huihui-Qwen3.6-35B-A3B-abliterated-ggml-model-PXQN3 scores 71.3/100 for coding. It runs at about 19.4 tok/s, so if you want more speed, byteshape/Qwen3.8-27B-GGUF:IQ4_XS-ASCII-Condensed is faster (~42.0 tok/s) and still scores well for coding (86.3/100).
Is Huihui-Qwen3.6-35B-A3B-abliterated-ggml-model-PXQN3 good for agentic (tool-using) tasks?
In our benchmarks, Huihui-Qwen3.6-35B-A3B-abliterated-ggml-model-PXQN3 scores 79.4/100 for agentic workflows. It runs at about 19.4 tok/s, so if you want more speed, granite-4.2-8b-Q4_K_L is faster (~71.6 tok/s) and still scores well for agentic workflows (92.9/100).
How fast is Huihui-Qwen3.6-35B-A3B-abliterated-ggml-model-PXQN3 for local inference?
Across 2 community benchmark runs, Huihui-Qwen3.6-35B-A3B-abliterated-ggml-model-PXQN3 reaches up to 20.9 tok/s and averages 19.4 tok/s, with the fastest results on NVIDIA GeForce GTX 1070.
Which tools have been used to run Huihui-Qwen3.6-35B-A3B-abliterated-ggml-model-PXQN3?
Benchmarks were submitted using llama.cpp. Results are community-contributed and updated as new runs arrive.