Model benchmarks
Qwen3.8NextO local LLM performance
As of September 2026, Qwen3.8NextO runs at up to 61.4 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
Unknown
Peak speed
61.4 tok/s
Average speed
58.2 tok/s
Avg PP
1799.1 tok/s
Min memory
70.1 GB
Max context
262.144 tokens
Avg output / run
14.321 tokens
Avg runtime / run
4m 24s
Avg quality
84.2
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.
| Task | P5 (low) | Avg | P95 (high) |
|---|---|---|---|
| Overall | 84.1 | 84.2 | 84.3 |
| Agent Workflow | 78.7 | 81.5 | 84.3 |
| Code Generation | 73.9 | 76.9 | 79.9 |
| Role Play & Narrative | 91.6 | 93.4 | 95.1 |
| Research & Analysis | 82.7 | 85.0 | 87.4 |
Performance by hardware and tool
Every hardware/tool/quantization combination Qwen3.8NextO has been benchmarked on, ranked by peak token generation speed. Last updated September 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| Apple M5 Max | oMLX | — | 61.4 tok/s | 58.2 tok/s | 70.1 GB | 262.144 tokens | 84.2 | 2 |
Benchmark runs
All 2 Qwen3.8NextO runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is Qwen3.8NextO good for coding?
- In our benchmarks, Qwen3.8NextO scores 76.9/100 for coding. It runs at about 58.2 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 Qwen3.8NextO good for agentic (tool-using) tasks?
- In our benchmarks, Qwen3.8NextO scores 81.5/100 for agentic workflows. It runs at about 58.2 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 Qwen3.8NextO for local inference?
- Across 2 community benchmark runs, Qwen3.8NextO reaches up to 61.4 tok/s and averages 58.2 tok/s, with the fastest results on Apple M5 Max.
- How much memory does Qwen3.8NextO need?
- The leanest observed configuration used about 70.1 GB of memory.
- Which tools have been used to run Qwen3.8NextO?
- Benchmarks were submitted using oMLX. Results are community-contributed and updated as new runs arrive.