Model benchmarks
Qwen3.8-27B-oQ4-mtp local LLM performance
As of August 2026, Qwen3.8-27B-oQ4-mtp runs at up to 47.7 tok/s for local inference (best of 4 community benchmark runs across 1 GPU).
Model size
27B
Peak speed
47.7 tok/s
Average speed
42.8 tok/s
Min memory
17.6 GB
Max context
262,144 tokens
Avg output / run
92,532 tokens
Avg runtime / run
19m 4s
Avg quality
83.9
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.
| Task | P5 (low) | Avg | P95 (high) |
|---|---|---|---|
| Overall | 81.2 | 83.9 | 87.0 |
| Agent Workflow | 86.5 | 89.1 | 90.7 |
| Code Generation | 70.5 | 75.4 | 79.6 |
| Role Play & Narrative | 71.4 | 85.0 | 95.1 |
| Research & Analysis | 80.6 | 86.1 | 88.9 |
Performance by hardware and tool
Every hardware/tool/quantization combination Qwen3.8-27B-oQ4-mtp has been benchmarked on, ranked by peak token generation speed. Last updated August 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| Apple M5 Max | oMLX | — | 47.7 tok/s | 42.8 tok/s | 17.6 GB | 262,144 tokens | 83.9 | 4 |
Benchmark runs
All 4 Qwen3.8-27B-oQ4-mtp runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is Qwen3.8-27B-oQ4-mtp good for coding?
- In our benchmarks, Qwen3.8-27B-oQ4-mtp scores 75.4/100 for coding. It runs at about 42.8 tok/s, so if you want more speed, Nex-N2.5-mini-IQ3_XXS is faster (~80.3 tok/s) and still scores well for coding (84.5/100).
- Is Qwen3.8-27B-oQ4-mtp good for agentic (tool-using) tasks?
- In our benchmarks, Qwen3.8-27B-oQ4-mtp scores 89.2/100 for agentic workflows. It runs at about 42.8 tok/s, so if you want more speed, Nex-N2.5-mini-APEX-Mini is faster (~88.2 tok/s) and still scores well for agentic workflows (91.3/100).
- How fast is Qwen3.8-27B-oQ4-mtp for local inference?
- Across 4 community benchmark runs, Qwen3.8-27B-oQ4-mtp reaches up to 47.7 tok/s and averages 42.8 tok/s, with the fastest results on Apple M5 Max.
- How much memory does Qwen3.8-27B-oQ4-mtp need?
- The leanest observed configuration used about 17.6 GB of memory.
- Which tools have been used to run Qwen3.8-27B-oQ4-mtp?
- Benchmarks were submitted using oMLX. Results are community-contributed and updated as new runs arrive.