Model benchmarks
Qwen3.8-Flash-Next-UD-IQ3_XXS local LLM performance
As of November 2026, Qwen3.8-Flash-Next-UD-IQ3_XXS runs at up to 140.9 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
Unknown
Peak speed
140.9 tok/s
Average speed
128.6 tok/s
Avg PP
2082.0 tok/s
Min memory
58.6 GB
Max context
65,536 tokens
Avg output / run
36,512 tokens
Avg runtime / run
4m 47s
Avg quality
76.0
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 | 75.4 | 76.0 | 76.6 |
| Agent Workflow | 78.0 | 85.6 | 93.1 |
| Code Generation | 24.4 | 45.9 | 67.4 |
| Role Play & Narrative | 79.8 | 87.7 | 95.6 |
| Research & Analysis | 81.3 | 84.9 | 88.5 |
Performance by hardware and tool
Every hardware/tool/quantization combination Qwen3.8-Flash-Next-UD-IQ3_XXS 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 | — | 140.9 tok/s | 128.6 tok/s | 58.6 GB | 65,536 tokens | 76.0 | 2 |
Benchmark runs
All 2 Qwen3.8-Flash-Next-UD-IQ3_XXS runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is Qwen3.8-Flash-Next-UD-IQ3_XXS good for coding?
- In our benchmarks, Qwen3.8-Flash-Next-UD-IQ3_XXS scores 45.9/100 for coding. It runs at about 128.6 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-UD-IQ3_XXS good for agentic (tool-using) tasks?
- In our benchmarks, Qwen3.8-Flash-Next-UD-IQ3_XXS scores 85.6/100 for agentic workflows. It runs at about 128.6 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-UD-IQ3_XXS for local inference?
- Across 2 community benchmark runs, Qwen3.8-Flash-Next-UD-IQ3_XXS reaches up to 140.9 tok/s and averages 128.6 tok/s, with the fastest results on Apple M5 Ultra.
- How much memory does Qwen3.8-Flash-Next-UD-IQ3_XXS need?
- The leanest observed configuration used about 58.6 GB of memory.
- Which tools have been used to run Qwen3.8-Flash-Next-UD-IQ3_XXS?
- Benchmarks were submitted using oMLX. Results are community-contributed and updated as new runs arrive.