Model benchmarks
qwen3.5:27b local LLM performance
As of July 2026, qwen3.5:27b runs at up to 10.5 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
27.8B
Peak speed
10.5 tok/s
Average speed
9.2 tok/s
Min memory
15.7 GB
Max context
8.192 tokens
Avg quality
60.6
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 | 58.2 | 60.6 | 62.9 |
| Agent Workflow | 67.3 | 76.3 | 85.2 |
| Code Generation | 4.8 | 16.8 | 28.7 |
| Role Play & Narrative | 76.8 | 78.9 | 80.9 |
| Research & Analysis | 65.9 | 70.4 | 74.8 |
Performance by hardware and tool
Every hardware/tool/quantization combination qwen3.5:27b has been benchmarked on, ranked by peak token generation speed. Last updated July 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| Apple M4 Pro | Ollama | Q4_K_M | 10.5 tok/s | 9.2 tok/s | 15.7 GB | 8.192 tokens | 60.6 | 2 |
Benchmark runs
All 2 qwen3.5:27b runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is qwen3.5:27b good for coding?
- In our benchmarks, qwen3.5:27b scores 16.8/100 for coding. It runs at about 9.2 tok/s, so if you want more speed, /home/polaris/models/Tiel-Coder-35B-A3B-MTP-UD-Q6_K_XL.gguf is faster (~35.6 tok/s) and still scores well for coding (84.9/100).
- Is qwen3.5:27b good for agentic (tool-using) tasks?
- In our benchmarks, qwen3.5:27b scores 76.3/100 for agentic workflows. It runs at about 9.2 tok/s, so if you want more speed, muse-glimmer:latest is faster (~34.5 tok/s) and still scores well for agentic workflows (93.0/100).
- How fast is qwen3.5:27b for local inference?
- Across 2 community benchmark runs, qwen3.5:27b reaches up to 10.5 tok/s and averages 9.2 tok/s, with the fastest results on Apple M4 Pro.
- How much memory does qwen3.5:27b need?
- The leanest observed configuration used about 15.7 GB of memory (quantizations tested: Q4_K_M).
- Which tools have been used to run qwen3.5:27b?
- Benchmarks were submitted using Ollama. Results are community-contributed and updated as new runs arrive.