Model benchmarks
IsValorum/Ornith-1_5-35B-A3B_APEX-I-MiniPlus-V2_1 local LLM performance
As of November 2026, IsValorum/Ornith-1_5-35B-A3B_APEX-I-MiniPlus-V2_1 runs at up to 23.1 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
35B
Peak speed
23.1 tok/s
Average speed
22.9 tok/s
Avg PP
277.9 tok/s
Min memory
8.2 GB
Max context
65,536 tokens
Avg output / run
27,685 tokens
Avg runtime / run
19m 30s
Avg quality
81.7
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 | 79.3 | 81.7 | 84.1 |
| Agent Workflow | 75.4 | 76.1 | 76.8 |
| Code Generation | 73.8 | 75.7 | 77.6 |
| Role Play & Narrative | 86.0 | 88.7 | 91.3 |
| Research & Analysis | 81.8 | 86.3 | 90.7 |
Performance by hardware and tool
Every hardware/tool/quantization combination IsValorum/Ornith-1_5-35B-A3B_APEX-I-MiniPlus-V2_1 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 3050 | llama.cpp | — | 23.1 tok/s | 22.9 tok/s | 8.2 GB | 65,536 tokens | 81.7 | 2 |
Benchmark runs
All 2 IsValorum/Ornith-1_5-35B-A3B_APEX-I-MiniPlus-V2_1 runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is IsValorum/Ornith-1_5-35B-A3B_APEX-I-MiniPlus-V2_1 good for coding?
- In our benchmarks, IsValorum/Ornith-1_5-35B-A3B_APEX-I-MiniPlus-V2_1 scores 75.7/100 for coding. It runs at about 22.9 tok/s, so if you want more speed, Qwen3.8-27B-Turbo-Brilliance-50X-RI-GSQ-RCO-IQ3_XXS-mtp is faster (~37.5 tok/s) and still scores well for coding (83.8/100).
- Is IsValorum/Ornith-1_5-35B-A3B_APEX-I-MiniPlus-V2_1 good for agentic (tool-using) tasks?
- In our benchmarks, IsValorum/Ornith-1_5-35B-A3B_APEX-I-MiniPlus-V2_1 scores 76.1/100 for agentic workflows. It runs at about 22.9 tok/s, so if you want more speed, muse-glimmer:latest is faster (~33.3 tok/s) and still scores well for agentic workflows (91.6/100).
- How fast is IsValorum/Ornith-1_5-35B-A3B_APEX-I-MiniPlus-V2_1 for local inference?
- Across 2 community benchmark runs, IsValorum/Ornith-1_5-35B-A3B_APEX-I-MiniPlus-V2_1 reaches up to 23.1 tok/s and averages 22.9 tok/s, with the fastest results on NVIDIA GeForce RTX 3050.
- How much memory does IsValorum/Ornith-1_5-35B-A3B_APEX-I-MiniPlus-V2_1 need?
- The leanest observed configuration used about 8.2 GB of memory.
- Which tools have been used to run IsValorum/Ornith-1_5-35B-A3B_APEX-I-MiniPlus-V2_1?
- Benchmarks were submitted using llama.cpp. Results are community-contributed and updated as new runs arrive.