Model benchmarks

qwen/qwen3.6-27b local LLM performance

As of July 2026, qwen/qwen3.6-27b runs at up to 38.3 tok/s for local inference (best of 3 community benchmark runs across 1 GPU).

LM Studio

Model size

27B

Peak speed

38.3 tok/s

Average speed

21.7 tok/s

Min memory

16.3 GB

Max context

7,138 tokens

Best quality

91.3

Benchmark runs

3

GPUs tested

1

Performance by hardware and tool

Every hardware/tool/quantization combination qwen/qwen3.6-27b has been benchmarked on, ranked by peak token generation speed. Last updated July 2026.

HardwareToolQuantBest tok/sAvg tok/sMemoryContextQualityRuns
AMD Radeon RX 7900 XTXLM Studio38.3 tok/s21.7 tok/s16.3 GB7,138 tokens82.23

Frequently asked questions

How fast is qwen/qwen3.6-27b for local inference?
Across 3 community benchmark runs, qwen/qwen3.6-27b reaches up to 38.3 tok/s and averages 21.7 tok/s, with the fastest results on AMD Radeon RX 7900 XTX.
How much memory does qwen/qwen3.6-27b need?
The leanest observed configuration used about 16.3 GB of memory.
Which tools have been used to run qwen/qwen3.6-27b?
Benchmarks were submitted using LM Studio. Results are community-contributed and updated as new runs arrive.