Model benchmarks
stablelm-2-zephyr-1_6b-Q8_0 local LLM performance
As of November 2026, stablelm-2-zephyr-1_6b-Q8_0 runs at up to 58.4 tok/s for local inference (best of 3 community benchmark runs across 1 GPU).
Model size
6B
Peak speed
58.4 tok/s
Average speed
58.0 tok/s
Avg PP
2697.7 tok/s
Min memory
n/a
Max context
65,536 tokens
Avg output / run
1,930 tokens
Avg runtime / run
37s
Avg quality
21.8
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 | 21.0 | 21.8 | 22.7 |
| Agent Workflow | 34.3 | 37.8 | 40.9 |
| Code Generation | 0.2 | 1.0 | 1.5 |
| Role Play & Narrative | 24.6 | 30.9 | 35.0 |
| Research & Analysis | 13.2 | 17.8 | 25.1 |
Performance by hardware and tool
Every hardware/tool/quantization combination stablelm-2-zephyr-1_6b-Q8_0 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 GeForce RTX 2050 | Lemonade | Q8_0 | 58.4 tok/s | 58.0 tok/s | n/a | 65,536 tokens | 21.8 | 3 |
Benchmark runs
All 3 stablelm-2-zephyr-1_6b-Q8_0 runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is stablelm-2-zephyr-1_6b-Q8_0 good for coding?
- In our benchmarks, stablelm-2-zephyr-1_6b-Q8_0 scores 1.0/100 for coding. It runs at about 58.0 tok/s, so if you want more speed, swift-1.5-27b is faster (~98.7 tok/s) and still scores well for coding (84.8/100).
- Is stablelm-2-zephyr-1_6b-Q8_0 good for agentic (tool-using) tasks?
- In our benchmarks, stablelm-2-zephyr-1_6b-Q8_0 scores 37.8/100 for agentic workflows. It runs at about 58.0 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 stablelm-2-zephyr-1_6b-Q8_0 for local inference?
- Across 3 community benchmark runs, stablelm-2-zephyr-1_6b-Q8_0 reaches up to 58.4 tok/s and averages 58.0 tok/s, with the fastest results on NVIDIA GeForce RTX 2050.
- Which tools have been used to run stablelm-2-zephyr-1_6b-Q8_0?
- Benchmarks were submitted using Lemonade. Results are community-contributed and updated as new runs arrive.