Model benchmarks
qwen2.5-coder:7b local LLM performance
As of October 2026, qwen2.5-coder:7b runs at up to 73.8 tok/s for local inference (best of 2 community benchmark runs across 2 GPUs).
Model size
7.6B
Peak speed
73.8 tok/s
Average speed
60.0 tok/s
Avg PP
1645.3 tok/s
Min memory
4.7 GB
Max context
8,192 tokens
Avg output / run
4,621 tokens
Avg runtime / run
1m 33s
Avg quality
30.8
Benchmark runs
2
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 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 | 30.3 | 30.8 | 31.3 |
| Agent Workflow | 16.6 | 22.8 | 29.0 |
| Code Generation | 25.8 | 31.7 | 37.5 |
| Role Play & Narrative | 28.2 | 29.4 | 30.5 |
| Research & Analysis | 36.8 | 39.4 | 42.1 |
Performance by hardware and tool
Every hardware/tool/quantization combination qwen2.5-coder:7b has been benchmarked on, ranked by peak token generation speed. Last updated October 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| NVIDIA GeForce RTX 3060 Ti | Ollama | Q4_K_M | 73.8 tok/s | 73.8 tok/s | 6.1 GB | 8,192 tokens | 30.3 | 1 |
| NVIDIA GeForce GTX 1080 | Ollama | Q4_K_M | 46.3 tok/s | 46.3 tok/s | 4.7 GB | 8,192 tokens | 31.3 | 1 |
Benchmark runs
All 2 qwen2.5-coder:7b runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is qwen2.5-coder:7b good for coding?
- In our benchmarks, qwen2.5-coder:7b scores 31.7/100 for coding. It runs at about 60.0 tok/s, so if you want more speed, Nex-N2.5-mini-IQ3_XXS is faster (~91.4 tok/s) and still scores well for coding (81.6/100).
- Is qwen2.5-coder:7b good for agentic (tool-using) tasks?
- In our benchmarks, qwen2.5-coder:7b scores 22.8/100 for agentic workflows. It runs at about 60.0 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 qwen2.5-coder:7b for local inference?
- Across 2 community benchmark runs, qwen2.5-coder:7b reaches up to 73.8 tok/s and averages 60.0 tok/s, with the fastest results on NVIDIA GeForce RTX 3060 Ti.
- How much memory does qwen2.5-coder:7b need?
- The leanest observed configuration used about 4.7 GB of memory (quantizations tested: Q4_K_M).
- Which tools have been used to run qwen2.5-coder:7b?
- Benchmarks were submitted using Ollama. Results are community-contributed and updated as new runs arrive.