Model benchmarks
Qwen3.8-9B-Distill-uncensored-heretic.i1-IQ3_M local LLM performance
As of September 2026, Qwen3.8-9B-Distill-uncensored-heretic.i1-IQ3_M runs at up to 34.3 tok/s for local inference (best of 3 community benchmark runs across 1 GPU).
Model size
9B
Peak speed
34.3 tok/s
Average speed
34.3 tok/s
Avg PP
295.1 tok/s
Min memory
4.4 GB
Max context
65.536 tokens
Avg output / run
7.060 tokens
Avg runtime / run
3m 36s
Avg quality
59.7
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 | 52.8 | 59.7 | 64.6 |
| Agent Workflow | 64.4 | 68.5 | 72.5 |
| Code Generation | 25.2 | 36.3 | 47.5 |
| Role Play & Narrative | 53.3 | 69.6 | 85.0 |
| Research & Analysis | 63.3 | 64.5 | 66.3 |
Performance by hardware and tool
Every hardware/tool/quantization combination Qwen3.8-9B-Distill-uncensored-heretic.i1-IQ3_M has been benchmarked on, ranked by peak token generation speed. Last updated September 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| Tesla P100-PCIE-16GB | llama.cpp | — | 34.3 tok/s | 34.3 tok/s | 4.4 GB | 65.536 tokens | 59.7 | 3 |
Benchmark runs
All 3 Qwen3.8-9B-Distill-uncensored-heretic.i1-IQ3_M runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is Qwen3.8-9B-Distill-uncensored-heretic.i1-IQ3_M good for coding?
- In our benchmarks, Qwen3.8-9B-Distill-uncensored-heretic.i1-IQ3_M scores 36.3/100 for coding. It runs at about 34.3 tok/s, so if you want more speed, unsloth/Qwen3.8-27B-GGUF:IQ3_S is faster (~104.3 tok/s) and still scores well for coding (83.0/100).
- Is Qwen3.8-9B-Distill-uncensored-heretic.i1-IQ3_M good for agentic (tool-using) tasks?
- In our benchmarks, Qwen3.8-9B-Distill-uncensored-heretic.i1-IQ3_M scores 68.5/100 for agentic workflows. It runs at about 34.3 tok/s, so if you want more speed, granite-4.2-8b-Q4_K_L is faster (~71.6 tok/s) and still scores well for agentic workflows (92.9/100).
- How fast is Qwen3.8-9B-Distill-uncensored-heretic.i1-IQ3_M for local inference?
- Across 3 community benchmark runs, Qwen3.8-9B-Distill-uncensored-heretic.i1-IQ3_M reaches up to 34.3 tok/s and averages 34.3 tok/s, with the fastest results on Tesla P100-PCIE-16GB.
- How much memory does Qwen3.8-9B-Distill-uncensored-heretic.i1-IQ3_M need?
- The leanest observed configuration used about 4.4 GB of memory.
- Which tools have been used to run Qwen3.8-9B-Distill-uncensored-heretic.i1-IQ3_M?
- Benchmarks were submitted using llama.cpp. Results are community-contributed and updated as new runs arrive.