1× DGX Spark
82.2tok/sMixed
1 request, mixed prompt set, with MTP (EAGLE/nextn) (per recipe) Source
Intelligenceno independent value
Local AI · Recipe · 4× RTX 3090
Ornith-1.5-35B-A3B EXL3 3 bpw with ExLlamaV3 on 4× RTX 3090
Ornith-1.5-35B-A3B EXL3 3 bpw with ExLlamaV3 on 4× RTX 3090: 50 tok/s according to huggingface.co (dataset as of Nov 3, 2026).
by 0xSero
Intelligence (original model) · Artificial Analysisno independent value
≤ 3 bit: quantization may cost quality
Engine
- Engine
- exllamav3 (TabbyAPI) 1.4.3 (+ Patch tp_import_split_n)
- Quantization
- EXL3 3 bpw Experten (MCG), Rest BF16, MTP 4 bpw
- Model family
- Ornith-1.5-35B-A3B
- Context
- 262,144
- Parameters
- 35B total / 3B active
- Creator
- 0xSero
- GitHub stars
- 0
- Repo updated
- Aug 26, 2026
Measurements
Every sourced value of this recipe, each with its condition and source. Bars relative to the largest value in the group.
Decode · 1 request
What you need
- Hardware
- 4 × NVIDIA RTX 3090
- Weights
- Main weights0xSero/Ornith-1.5-35B-A3B-EXL3-3bpw MIT
- Engine
- exllamav3 (TabbyAPI) 1.4.3 (+ Patch tp_import_split_n)
- Context
- 262,144 tokens
Notes
What matters before you rebuild it.
- ≤ 3 bit: quantization may cost qualityAt 3 bit and below the model may answer noticeably worse than the original. The intelligence number refers to the original.
Sources
Related recipes
4× RTX 3090
145tok/sEveryday
1 request, prose prompt, with DFlash (per recipe) Source
Intelligence34with thinkingno thinking 20
- Experimental
4× RTX 3090
110tok/sEveryday
1 request, prose prompt, with MTP (per recipe) Source
Intelligence40with thinking
- License unclear
4× RTX 3090
105tok/sEveryday
1 request, prose prompt, with MTP (per recipe) Source
Intelligence21with thinkingno thinking 20
4× RTX 3090
104tok/sEveryday
1 request, prose prompt, with MTP (per recipe) Source
Intelligenceno independent value
- Experimental
- License unclear
4× RTX 3090
85.3tok/sEveryday
1 request, prose prompt, with DFlash2 (per recipe) Source
Intelligence34with thinkingno thinking 20
Local AI in your company?
In a workshop we work out which models and which hardware fit your tasks, and build the first agent on your infrastructure.