Model benchmarks
qwen3.8-9b-distill local LLM performance
As of November 2026, qwen3.8-9b-distill runs at up to 119.5 tok/s for local inference (best of 3 community benchmark runs across 1 GPU).
Model size
9B
Peak speed
119.5 tok/s
Average speed
115.6 tok/s
Avg PP
3051.1 tok/s
Min memory
n/a
Max context
65,536 tokens
Avg output / run
10,953 tokens
Avg runtime / run
1m 33s
Avg quality
69.3
Benchmark runs
3
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 3 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 | 63.6 | 69.3 | 73.1 |
| Agent Workflow | 43.4 | 60.7 | 75.4 |
| Code Generation | 51.0 | 58.2 | 64.5 |
| Role Play & Narrative | 82.5 | 84.3 | 85.9 |
| Research & Analysis | 72.2 | 73.8 | 75.8 |
Performance by hardware and tool
Every hardware/tool/quantization combination qwen3.8-9b-distill 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 |
|---|---|---|---|---|---|---|---|---|
| NVIDIA A10G | ninfer | — | 119.5 tok/s | 115.6 tok/s | n/a | 65,536 tokens | 69.3 | 3 |
Benchmark runs
All 3 qwen3.8-9b-distill runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is qwen3.8-9b-distill good for coding?
- In our benchmarks, qwen3.8-9b-distill scores 58.2/100 for coding. It runs at about 115.6 tok/s, so if you want more speed, Qwen3.6-35B-A3B-Q4_K_XL is faster (~156.9 tok/s) and still scores well for coding (82.2/100).
- Is qwen3.8-9b-distill good for agentic (tool-using) tasks?
- In our benchmarks, qwen3.8-9b-distill scores 60.7/100 for agentic workflows. It runs at about 115.6 tok/s, so if you want more speed, swift-1.5-qwen3.8-flash-next-iq3_s is faster (~259.0 tok/s) and still scores well for agentic workflows (88.9/100).
- How fast is qwen3.8-9b-distill for local inference?
- Across 3 community benchmark runs, qwen3.8-9b-distill reaches up to 119.5 tok/s and averages 115.6 tok/s, with the fastest results on NVIDIA A10G.
- Which tools have been used to run qwen3.8-9b-distill?
- Benchmarks were submitted using ninfer. Results are community-contributed and updated as new runs arrive.