Model benchmarks
Ornith-1.5-9B-Q8_0 local LLM performance
As of September 2026, Ornith-1.5-9B-Q8_0 runs at up to 71.9 tok/s for local inference (best of 3 community benchmark runs across 2 GPUs).
Model size
9B
Peak speed
71.9 tok/s
Average speed
63.4 tok/s
Avg PP
1249.6 tok/s
Min memory
n/a
Max context
131,072 tokens
Avg output / run
58,340 tokens
Avg runtime / run
14m 17s
Avg quality
64.4
Benchmark runs
3
GPUs tested
2
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 | 56.2 | 64.4 | 74.8 |
| Agent Workflow | 8.4 | 26.4 | 56.0 |
| Code Generation | 48.9 | 61.2 | 70.3 |
| Role Play & Narrative | 82.3 | 86.0 | 89.8 |
| Research & Analysis | 83.2 | 84.0 | 84.6 |
Performance by hardware and tool
Every hardware/tool/quantization combination Ornith-1.5-9B-Q8_0 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 |
|---|---|---|---|---|---|---|---|---|
| NVIDIA GeForce RTX 5060 | llama.cpp | Q8_0 | 71.9 tok/s | 67.7 tok/s | n/a | 131,072 tokens | 68.7 | 2 |
| AMD Radeon RX 9070/9070 XT/9070 GRE | vLLM | Q8_0 | 54.9 tok/s | 54.9 tok/s | n/a | 8,192 tokens | 55.7 | 1 |
Benchmark runs
All 3 Ornith-1.5-9B-Q8_0 runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is Ornith-1.5-9B-Q8_0 good for coding?
- In our benchmarks, Ornith-1.5-9B-Q8_0 scores 61.2/100 for coding. It runs at about 63.4 tok/s, so if you want more speed, ornith-ai/Ornith-1.5-35B-A3B-GGUF is faster (~90.8 tok/s) and still scores well for coding (83.8/100).
- Is Ornith-1.5-9B-Q8_0 good for agentic (tool-using) tasks?
- In our benchmarks, Ornith-1.5-9B-Q8_0 scores 26.4/100 for agentic workflows. It runs at about 63.4 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 Ornith-1.5-9B-Q8_0 for local inference?
- Across 3 community benchmark runs, Ornith-1.5-9B-Q8_0 reaches up to 71.9 tok/s and averages 63.4 tok/s, with the fastest results on NVIDIA GeForce RTX 5060.
- Which tools have been used to run Ornith-1.5-9B-Q8_0?
- Benchmarks were submitted using llama.cpp, vLLM. Results are community-contributed and updated as new runs arrive.