Model benchmarks
qwen3:14b local LLM performance
As of August 2026, qwen3:14b runs at up to 33.4 tok/s for local inference (best of 4 community benchmark runs across 2 GPUs).
OllamaQ4_K_M
Model size
14.8B
Peak speed
33.4 tok/s
Average speed
18.7 tok/s
Min memory
10.0 GB
Max context
196,608 tokens
Avg runtime / run
13m 8s
Avg quality
59.4
Benchmark runs
4
GPUs tested
2
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 | 51.7 | 59.4 | 63.8 |
| Agent Workflow | 56.3 | 71.4 | 80.4 |
| Code Generation | 4.0 | 30.6 | 54.4 |
| Role Play & Narrative | 71.6 | 76.2 | 79.5 |
| Research & Analysis | 50.0 | 59.2 | 72.7 |
Performance by hardware and tool
Every hardware/tool/quantization combination qwen3:14b 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 |
|---|---|---|---|---|---|---|---|---|
| AMD Radeon RX 7700 XT / 7800 XT | Ollama | Q4_K_M | 33.4 tok/s | 33.4 tok/s | 15.2 GB | 196,608 tokens | 64.2 | 1 |
| AMD Radeon RX 7900 XTX | Ollama | Q4_K_M | 16.5 tok/s | 13.7 tok/s | 10.0 GB | 65,536 tokens | 57.7 | 3 |
Frequently asked questions
- Is qwen3:14b good for coding?
- In our benchmarks, qwen3:14b scores 30.6/100 for coding. It runs at about 18.7 tok/s, so if you want more speed, Gemma4:E2B/QAT-MTP@131K is faster (~303.9 tok/s) and still scores well for coding (66.9/100).
- Is qwen3:14b good for agentic (tool-using) tasks?
- In our benchmarks, qwen3:14b scores 71.5/100 for agentic workflows. It runs at about 18.7 tok/s, so if you want more speed, Gemma4:E2B/QAT-MTP@131K is faster (~303.9 tok/s) and still scores well for agentic workflows (71.7/100).
- How fast is qwen3:14b for local inference?
- Across 4 community benchmark runs, qwen3:14b reaches up to 33.4 tok/s and averages 18.7 tok/s, with the fastest results on AMD Radeon RX 7700 XT / 7800 XT.
- How much memory does qwen3:14b need?
- The leanest observed configuration used about 10.0 GB of memory (quantizations tested: Q4_K_M).
- Which tools have been used to run qwen3:14b?
- Benchmarks were submitted using Ollama. Results are community-contributed and updated as new runs arrive.