Model benchmarks
ornith15-heretic-mini local LLM performance
As of October 2026, ornith15-heretic-mini runs at up to 73.6 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
Unknown
Peak speed
73.6 tok/s
Average speed
72.9 tok/s
Avg PP
759.3 tok/s
Min memory
20.4 GB
Max context
65,536 tokens
Avg output / run
8,197 tokens
Avg runtime / run
2m 2s
Avg quality
76.1
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 | 74.6 | 76.1 | 77.5 |
| Agent Workflow | 79.4 | 80.7 | 82.1 |
| Code Generation | 52.3 | 59.0 | 65.6 |
| Role Play & Narrative | 85.7 | 86.6 | 87.5 |
| Research & Analysis | 75.1 | 78.1 | 81.1 |
Performance by hardware and tool
Every hardware/tool/quantization combination ornith15-heretic-mini 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 |
|---|---|---|---|---|---|---|---|---|
| NVIDIA RTX A2000 12GB | llama.cpp | — | 73.6 tok/s | 72.9 tok/s | 20.4 GB | 65,536 tokens | 76.1 | 2 |
Benchmark runs
All 2 ornith15-heretic-mini runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is ornith15-heretic-mini good for coding?
- In our benchmarks, ornith15-heretic-mini scores 59.0/100 for coding. It runs at about 72.9 tok/s, so if you want more speed, Nex-N2.5-mini-IQ3_XXS is faster (~91.4 tok/s) and still scores well for coding (81.6/100).
- Is ornith15-heretic-mini good for agentic (tool-using) tasks?
- In our benchmarks, ornith15-heretic-mini scores 80.7/100 for agentic workflows. It runs at about 72.9 tok/s, so if you want more speed, Nex-N2.5-mini-APEX-Mini is faster (~88.2 tok/s) and still scores well for agentic workflows (91.3/100).
- How fast is ornith15-heretic-mini for local inference?
- Across 2 community benchmark runs, ornith15-heretic-mini reaches up to 73.6 tok/s and averages 72.9 tok/s, with the fastest results on NVIDIA RTX A2000 12GB.
- How much memory does ornith15-heretic-mini need?
- The leanest observed configuration used about 20.4 GB of memory.
- Which tools have been used to run ornith15-heretic-mini?
- Benchmarks were submitted using llama.cpp. Results are community-contributed and updated as new runs arrive.