Model benchmarks
qwen3.8-2b-distill local LLM performance
As of October 2026, qwen3.8-2b-distill runs at up to 15.8 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
2B
Peak speed
15.8 tok/s
Average speed
15.2 tok/s
Avg PP
593.9 tok/s
Min memory
13.3 GB
Max context
8,192 tokens
Avg output / run
10,155 tokens
Avg runtime / run
10m 14s
Avg quality
33.3
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 | 24.0 | 33.3 | 42.7 |
| Agent Workflow | 25.8 | 54.3 | 82.7 |
| Code Generation | 8.3 | 13.2 | 18.1 |
| Role Play & Narrative | 32.4 | 33.8 | 35.1 |
| Research & Analysis | 26.9 | 32.2 | 37.4 |
Performance by hardware and tool
Every hardware/tool/quantization combination qwen3.8-2b-distill 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 Vega Series / Radeon Vega Mobile Series | Bionic | — | 15.8 tok/s | 15.2 tok/s | 13.3 GB | 8,192 tokens | 33.3 | 2 |
Benchmark runs
All 2 qwen3.8-2b-distill runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is qwen3.8-2b-distill good for coding?
- In our benchmarks, qwen3.8-2b-distill scores 13.2/100 for coding. It runs at about 15.2 tok/s, so if you want more speed, Qwen3.8-27B-oQ4e-fp16-mtp is faster (~38.6 tok/s) and still scores well for coding (82.6/100).
- Is qwen3.8-2b-distill good for agentic (tool-using) tasks?
- In our benchmarks, qwen3.8-2b-distill scores 54.3/100 for agentic workflows. It runs at about 15.2 tok/s, so if you want more speed, muse-glimmer:latest is faster (~33.3 tok/s) and still scores well for agentic workflows (91.6/100).
- How fast is qwen3.8-2b-distill for local inference?
- Across 2 community benchmark runs, qwen3.8-2b-distill reaches up to 15.8 tok/s and averages 15.2 tok/s, with the fastest results on AMD Radeon Vega Series / Radeon Vega Mobile Series.
- How much memory does qwen3.8-2b-distill need?
- The leanest observed configuration used about 13.3 GB of memory.
- Which tools have been used to run qwen3.8-2b-distill?
- Benchmarks were submitted using Bionic. Results are community-contributed and updated as new runs arrive.