Model benchmarks
qwen2.5:14b local LLM performance
As of November 2026, qwen2.5:14b runs at up to 10.3 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
14B
Peak speed
10.3 tok/s
Average speed
8.1 tok/s
Avg PP
465.7 tok/s
Min memory
8.6 GB
Max context
16,384 tokens
Avg output / run
2,488 tokens
Avg runtime / run
6m 48s
Avg quality
28.8
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 | 2.9 | 28.8 | 54.6 |
| Agent Workflow | 3.3 | 33.3 | 63.2 |
| Code Generation | 2.0 | 19.6 | 37.1 |
| Role Play & Narrative | 3.2 | 31.7 | 60.3 |
| Research & Analysis | 3.0 | 30.5 | 57.9 |
Performance by hardware and tool
Every hardware/tool/quantization combination qwen2.5:14b 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 |
|---|---|---|---|---|---|---|---|---|
| Intel Graphics | Ollama | — | 10.3 tok/s | 10.3 tok/s | 8.6 GB | 8,192 tokens | 57.5 | 1 |
| Intel Graphics | Ollama | Q4_K_M | 5.9 tok/s | 5.9 tok/s | 14.3 GB | 16,384 tokens | 0.0 | 1 |
Benchmark runs
All 2 qwen2.5:14b runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is qwen2.5:14b good for coding?
- In our benchmarks, qwen2.5:14b scores 19.6/100 for coding. It runs at about 8.1 tok/s, so if you want more speed, tensorfold-qwen38-27b is faster (~25.0 tok/s) and still scores well for coding (84.3/100).
- Is qwen2.5:14b good for agentic (tool-using) tasks?
- In our benchmarks, qwen2.5:14b scores 33.3/100 for agentic workflows. It runs at about 8.1 tok/s, so if you want more speed, muse-glimmer:latest is faster (~33.3 tok/s) and still scores well for agentic workflows (91.6/100).
- How fast is qwen2.5:14b for local inference?
- Across 2 community benchmark runs, qwen2.5:14b reaches up to 10.3 tok/s and averages 8.1 tok/s, with the fastest results on Intel Graphics.
- How much memory does qwen2.5:14b need?
- The leanest observed configuration used about 8.6 GB of memory (quantizations tested: Q4_K_M).
- Which tools have been used to run qwen2.5:14b?
- Benchmarks were submitted using Ollama. Results are community-contributed and updated as new runs arrive.