Model benchmarks

ornith:35b local LLM performance

As of July 2026, ornith:35b runs at up to 56.4 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).

OllamaQ4_K_M

Model size

34.7B

Peak speed

56.4 tok/s

Average speed

45.9 tok/s

Min memory

20.9 GB

Max context

5,628 tokens

Best quality

89.9

Benchmark runs

2

GPUs tested

1

Performance by hardware and tool

Every hardware/tool/quantization combination ornith:35b has been benchmarked on, ranked by peak token generation speed. Last updated July 2026.

HardwareToolQuantBest tok/sAvg tok/sMemoryContextQualityRuns
AMD Radeon RX 7900 XTXOllamaQ4_K_M56.4 tok/s45.9 tok/s20.9 GB5,628 tokens72.82

Frequently asked questions

How fast is ornith:35b for local inference?
Across 2 community benchmark runs, ornith:35b reaches up to 56.4 tok/s and averages 45.9 tok/s, with the fastest results on AMD Radeon RX 7900 XTX.
How much memory does ornith:35b need?
The leanest observed configuration used about 20.9 GB of memory (quantizations tested: Q4_K_M).
Which tools have been used to run ornith:35b?
Benchmarks were submitted using Ollama. Results are community-contributed and updated as new runs arrive.