Model benchmarks
Qwen3.8-Flash-Next-oQ3.5e-mtp local LLM performance
As of November 2026, Qwen3.8-Flash-Next-oQ3.5e-mtp runs at up to 99.6 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
125B
Peak speed
99.6 tok/s
Average speed
99.1 tok/s
Avg PP
874.0 tok/s
Min memory
63.2 GB
Max context
262,144 tokens
Avg output / run
82,006 tokens
Avg runtime / run
13m 34s
Avg quality
86.2
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 | 82.4 | 86.2 | 89.9 |
| Agent Workflow | 88.9 | 91.0 | 93.1 |
| Code Generation | 61.0 | 72.0 | 83.0 |
| Role Play & Narrative | 89.0 | 92.5 | 96.0 |
| Research & Analysis | 87.5 | 89.1 | 90.8 |
Performance by hardware and tool
Every hardware/tool/quantization combination Qwen3.8-Flash-Next-oQ3.5e-mtp has been benchmarked on, ranked by peak token generation speed. Last updated November 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| Apple M5 Ultra | oMLX | oQ3 | 99.6 tok/s | 99.1 tok/s | 63.2 GB | 262,144 tokens | 86.2 | 2 |
Benchmark runs
All 2 Qwen3.8-Flash-Next-oQ3.5e-mtp runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is Qwen3.8-Flash-Next-oQ3.5e-mtp good for coding?
- In our benchmarks, Qwen3.8-Flash-Next-oQ3.5e-mtp scores 72.0/100 for coding. It runs at about 99.1 tok/s, so if you want more speed, Qwen3.6-35B-A3B-Q4_K_XL is faster (~156.9 tok/s) and still scores well for coding (82.2/100).
- Is Qwen3.8-Flash-Next-oQ3.5e-mtp good for agentic (tool-using) tasks?
- In our benchmarks, Qwen3.8-Flash-Next-oQ3.5e-mtp scores 91.0/100 for agentic workflows. It runs at about 99.1 tok/s, so if you want more speed, swift-1.5-qwen3.8-flash-next-iq3_s is faster (~259.0 tok/s) and still scores well for agentic workflows (88.9/100).
- How fast is Qwen3.8-Flash-Next-oQ3.5e-mtp for local inference?
- Across 2 community benchmark runs, Qwen3.8-Flash-Next-oQ3.5e-mtp reaches up to 99.6 tok/s and averages 99.1 tok/s, with the fastest results on Apple M5 Ultra.
- How much memory does Qwen3.8-Flash-Next-oQ3.5e-mtp need?
- The leanest observed configuration used about 63.2 GB of memory (quantizations tested: oQ3).
- Which tools have been used to run Qwen3.8-Flash-Next-oQ3.5e-mtp?
- Benchmarks were submitted using oMLX. Results are community-contributed and updated as new runs arrive.