Model benchmarks
qwen2.5-coder:7b-instruct local LLM performance
As of July 2026, qwen2.5-coder:7b-instruct runs at up to 17.2 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
OllamaQ4_K_M
Model size
7.6B
Peak speed
17.2 tok/s
Average speed
15.2 tok/s
Min memory
4.6 GB
Max context
8,192 tokens
Avg quality
44.6
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 | 38.6 | 44.6 | 50.6 |
| Agent Workflow | 44.7 | 57.4 | 70.2 |
| Code Generation | 25.4 | 38.1 | 50.7 |
| Role Play & Narrative | 49.9 | 54.9 | 59.8 |
| Research & Analysis | 24.4 | 28.1 | 31.7 |
Performance by hardware and tool
Every hardware/tool/quantization combination qwen2.5-coder:7b-instruct 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 | Ollama | Q4_K_M | 17.2 tok/s | 15.2 tok/s | 4.6 GB | 8,192 tokens | 44.6 | 2 |
Frequently asked questions
- Is qwen2.5-coder:7b-instruct good for coding?
- In our benchmarks, qwen2.5-coder:7b-instruct scores 38.1/100 for coding. It runs at about 15.2 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 qwen2.5-coder:7b-instruct good for agentic (tool-using) tasks?
- In our benchmarks, qwen2.5-coder:7b-instruct scores 57.4/100 for agentic workflows. It runs at about 15.2 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 qwen2.5-coder:7b-instruct for local inference?
- Across 2 community benchmark runs, qwen2.5-coder:7b-instruct reaches up to 17.2 tok/s and averages 15.2 tok/s, with the fastest results on Apple M4.
- How much memory does qwen2.5-coder:7b-instruct need?
- The leanest observed configuration used about 4.6 GB of memory (quantizations tested: Q4_K_M).
- Which tools have been used to run qwen2.5-coder:7b-instruct?
- Benchmarks were submitted using Ollama. Results are community-contributed and updated as new runs arrive.