community pool / model answer

Ornith-1.5-35B-A3B-GGUF

chat · ornith-ai/Ornith-1.5-35B-A3B-GGUF

42.7 tok/s · model median · 9 runs in the model record.

Reference rig	Quantization	Median result	Basis	TTFT (ms)	Peak VRAM	Max context	Engines
Arc Pro B70 32GB	4-bit	131.5 tok/s	reported	153	- GB	8,192	llama.cpp
RTX 3080 12GB	4-bit	60.8 tok/s	reported	4,184	9.6 GB	131,072	llama.cpp
RTX 4070 12GB	4-bit	33.5 tok/s	measured	650	10.2 GB	262,144	llama.cpp

Context tested

The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.