Model benchmarks
qwen3-coder:latest local LLM performance
As of October 2026, qwen3-coder:latest runs at up to 96.0 tok/s for local inference (best of 5 community benchmark runs across 2 GPUs).
Model size
30.5B
Peak speed
96.0 tok/s
Average speed
34.3 tok/s
Avg PP
240.4 tok/s
Min memory
18.3 GB
Max context
65,536 tokens
Avg output / run
6,812 tokens
Avg runtime / run
8m 41s
Avg quality
63.5
Benchmark runs
5
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 5 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 | 57.6 | 63.5 | 70.5 |
| Agent Workflow | 51.0 | 63.5 | 85.1 |
| Code Generation | 58.5 | 63.4 | 66.8 |
| Role Play & Narrative | 61.0 | 69.6 | 76.7 |
| Research & Analysis | 50.5 | 57.6 | 66.7 |
Performance by hardware and tool
Every hardware/tool/quantization combination qwen3-coder:latest 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 |
|---|---|---|---|---|---|---|---|---|
| AMD Radeon RX 7900 XTX | Ollama | Q4_K_M | 96.0 tok/s | 64.6 tok/s | 18.9 GB | 65,536 tokens | 64.3 | 2 |
| Intel Arc B390 | Ollama | Q4_K_M | 15.0 tok/s | 14.1 tok/s | 18.3 GB | 32,768 tokens | 63.0 | 3 |
Benchmark runs
All 5 qwen3-coder:latest runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is qwen3-coder:latest good for coding?
- In our benchmarks, qwen3-coder:latest scores 63.4/100 for coding. It runs at about 34.3 tok/s, so if you want more speed, nvidia/Qwen3.8-27B-NVFP4 is faster (~63.3 tok/s) and still scores well for coding (82.7/100).
- Is qwen3-coder:latest good for agentic (tool-using) tasks?
- In our benchmarks, qwen3-coder:latest scores 63.5/100 for agentic workflows. It runs at about 34.3 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 qwen3-coder:latest for local inference?
- Across 5 community benchmark runs, qwen3-coder:latest reaches up to 96.0 tok/s and averages 34.3 tok/s, with the fastest results on AMD Radeon RX 7900 XTX.
- How much memory does qwen3-coder:latest need?
- The leanest observed configuration used about 18.3 GB of memory (quantizations tested: Q4_K_M).
- Which tools have been used to run qwen3-coder:latest?
- Benchmarks were submitted using Ollama. Results are community-contributed and updated as new runs arrive.