Model benchmarks
Ornith-1.5-35B-A3B-GSQ-RCO-3bit local LLM performance
As of September 2026, Ornith-1.5-35B-A3B-GSQ-RCO-3bit runs at up to 104.3 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
35B
Peak speed
104.3 tok/s
Average speed
103.0 tok/s
Avg PP
939.7 tok/s
Min memory
17.1 GB
Max context
131.072 tokens
Avg output / run
32.084 tokens
Avg runtime / run
5m 21s
Avg quality
84.3
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 | 83.8 | 84.3 | 84.7 |
| Agent Workflow | 83.0 | 86.5 | 90.0 |
| Code Generation | 72.2 | 75.6 | 79.0 |
| Role Play & Narrative | 89.4 | 90.0 | 90.7 |
| Research & Analysis | 82.7 | 85.0 | 87.3 |
Performance by hardware and tool
Every hardware/tool/quantization combination Ornith-1.5-35B-A3B-GSQ-RCO-3bit 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 | — | 104.3 tok/s | 103.0 tok/s | 17.1 GB | 131.072 tokens | 84.3 | 2 |
Benchmark runs
All 2 Ornith-1.5-35B-A3B-GSQ-RCO-3bit runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is Ornith-1.5-35B-A3B-GSQ-RCO-3bit good for coding?
- In our benchmarks, Ornith-1.5-35B-A3B-GSQ-RCO-3bit scores 75.6/100 for coding. It runs at about 103.0 tok/s, so if you want more speed, Qwen3.6-35B-A3B-Q4_K_XL is faster (~156.9 tok/s) and still scores well for coding (82.2/100).
- Is Ornith-1.5-35B-A3B-GSQ-RCO-3bit good for agentic (tool-using) tasks?
- In our benchmarks, Ornith-1.5-35B-A3B-GSQ-RCO-3bit scores 86.5/100 for agentic workflows. It runs at about 103.0 tok/s, so if you want more speed, Tiel-Coder-35B-A3B-Q4_K_S is faster (~162.4 tok/s) and still scores well for agentic workflows (87.8/100).
- How fast is Ornith-1.5-35B-A3B-GSQ-RCO-3bit for local inference?
- Across 2 community benchmark runs, Ornith-1.5-35B-A3B-GSQ-RCO-3bit reaches up to 104.3 tok/s and averages 103.0 tok/s, with the fastest results on NVIDIA GeForce RTX 5060.
- How much memory does Ornith-1.5-35B-A3B-GSQ-RCO-3bit need?
- The leanest observed configuration used about 17.1 GB of memory.
- Which tools have been used to run Ornith-1.5-35B-A3B-GSQ-RCO-3bit?
- Benchmarks were submitted using llama.cpp. Results are community-contributed and updated as new runs arrive.