Model benchmarks

Qwen3.8Next local LLM performance

As of September 2026, Qwen3.8Next runs at up to 70.4 tok/s for local inference (best of 4 community benchmark runs across 1 GPU).

oMLX
ShareRedditX

Model size

Unknown

Peak speed

70.4 tok/s

Average speed

54.7 tok/s

Avg PP

935.2 tok/s

Min memory

70.3 GB

Max context

262.144 tokens

Avg output / run

13.214 tokens

Avg runtime / run

4m 39s

Avg quality

83.0

Benchmark runs

4

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 4 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)
Overall78.983.085.4
Agent Workflow78.484.790.4
Code Generation68.874.781.3
Role Play & Narrative84.589.492.5
Research & Analysis81.083.385.6

Performance by hardware and tool

Every hardware/tool/quantization combination Qwen3.8Next has been benchmarked on, ranked by peak token generation speed. Last updated September 2026.

HardwareToolQuantBest tok/sAvg tok/sMemoryContextQualityRuns
Apple M5 MaxoMLX—70.4 tok/s54.7 tok/s70.3 GB262.144 tokens83.04

Benchmark runs

All 4 Qwen3.8Next runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.

Frequently asked questions

Is Qwen3.8Next good for coding?
In our benchmarks, Qwen3.8Next scores 74.7/100 for coding. It runs at about 54.7 tok/s, so if you want more speed, ornith-ai/Ornith-1.5-35B-A3B-GGUF is faster (~96.3 tok/s) and still scores well for coding (83.4/100).
Is Qwen3.8Next good for agentic (tool-using) tasks?
In our benchmarks, Qwen3.8Next scores 84.7/100 for agentic workflows. It runs at about 54.7 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 Qwen3.8Next for local inference?
Across 4 community benchmark runs, Qwen3.8Next reaches up to 70.4 tok/s and averages 54.7 tok/s, with the fastest results on Apple M5 Max.
How much memory does Qwen3.8Next need?
The leanest observed configuration used about 70.3 GB of memory.
Which tools have been used to run Qwen3.8Next?
Benchmarks were submitted using oMLX. Results are community-contributed and updated as new runs arrive.