Model benchmarks
r3060 local LLM performance
As of November 2026, r3060 runs at up to 101.8 tok/s for local inference (best of 5 community benchmark runs across 1 GPU).
Model size
Unknown
Peak speed
101.8 tok/s
Average speed
50.3 tok/s
Avg PP
940.4 tok/s
Min memory
n/a
Max context
65,536 tokens
Avg output / run
17,881 tokens
Avg runtime / run
7m 13s
Avg quality
74.7
Benchmark runs
5
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 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 | 71.3 | 74.7 | 76.9 |
| Agent Workflow | 64.6 | 76.0 | 87.5 |
| Code Generation | 64.5 | 68.3 | 72.3 |
| Role Play & Narrative | 78.3 | 82.0 | 84.7 |
| Research & Analysis | 68.1 | 72.2 | 77.0 |
Performance by hardware and tool
Every hardware/tool/quantization combination r3060 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 |
|---|---|---|---|---|---|---|---|---|
| Quadro K620 | llama.cpp | — | 101.8 tok/s | 50.3 tok/s | n/a | 65,536 tokens | 74.6 | 5 |
Benchmark runs
All 5 r3060 runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is r3060 good for coding?
- In our benchmarks, r3060 scores 68.3/100 for coding. It runs at about 50.3 tok/s, so if you want more speed, turbo is faster (~124.1 tok/s) and still scores well for coding (85.3/100).
- Is r3060 good for agentic (tool-using) tasks?
- In our benchmarks, r3060 scores 76.1/100 for agentic workflows. It runs at about 50.3 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 r3060 for local inference?
- Across 5 community benchmark runs, r3060 reaches up to 101.8 tok/s and averages 50.3 tok/s, with the fastest results on Quadro K620.
- Which tools have been used to run r3060?
- Benchmarks were submitted using llama.cpp. Results are community-contributed and updated as new runs arrive.