1× DGX Spark
60.3tok/sEveryday
1 request, prose prompt (essay), context 2k, no drafter (per recipe) Source
Intelligenceno independent value
- Custom kernel required
Local AI · Creator
vcruz305
25 published recipes.
Recipes
1× DGX Spark
53.3tok/sEveryday
1 request, prose prompt, with MTP (per recipe) Source
Intelligence40with thinking
- Custom kernel required
- License unclear
4× DGX Spark
35.7tok/sEveryday
1 request, prose prompt, with DFlash2, thinking on (per recipe) Source
Intelligence45with thinking
- Custom kernel required
- Non-commercial
4× DGX Spark
30.2tok/sEveryday
1 request, prose prompt, with DSpark (per recipe) Source
Intelligence40with thinkingno thinking 25
- Custom kernel required
1× DGX Spark
24.4tok/sEveryday
1 request, prose prompt, with DSpark (per recipe) Source
Intelligenceno independent value
- ≤ 3 bit: quantization may cost quality
- Custom kernel required
- License unclear
1× DGX Spark
~20.1tok/sEverydayapproximate · method not stated
1 request, chat prompt (1200-token replies), sampling 0.6/0.95, with DSpark (per recipe), range 17.7–22.5 tok/s Source
Intelligenceno independent value
- ≤ 3 bit: quantization may cost quality
- Custom kernel required
1× DGX Spark
17.6tok/sEveryday
1 request, realistic prompt, context 2k, with DFlash2 (per recipe) Source
Intelligence42with thinking
- ≤ 3 bit: quantization may cost quality
- Custom kernel required
- Non-commercial
3× DGX Spark
6.59tok/sEveryday
1 request, prose prompt, no speculative decoding Source
Intelligence44with thinkingOriginal model's value. Heavily compressed (~0.95 bit), intelligence is likely much lower.
- ≤ 3 bit: quantization may cost quality
- Experimental
- Very heavy quantization
- License unclear
1× DGX Spark
113tok/sMixed
1 request, mixed prompt set, no speculative decoding Source
Intelligence13with thinking
- ≤ 3 bit: quantization may cost quality
- License unclear
1× Strix Halo
41.3tok/sMixed
1 request, mixed prompt set, with ndt=3 Source
Intelligence40with thinking
- Custom kernel required
- License unclear
1× DGX Spark
17.5tok/sMixed
1 request, mixed prompt set, with DSpark Source
Intelligence40with thinkingno thinking 25Original model's value. Heavily compressed (~1.6 bit), intelligence is likely much lower.
- ≤ 3 bit: quantization may cost quality
- Very heavy quantization
- Custom kernel required
1× DGX Spark
17.3tok/sMixed
1 request, mixed prompt set, context 64k, with MTP Source
Intelligence42with thinkingOriginal model's value. Heavily compressed (~2.4 bit), intelligence is likely much lower.
- ≤ 3 bit: quantization may cost quality
- Custom kernel required
1× DGX Spark
17.2tok/sMixed
1 request, mixed prompt set, context 64k, with MTP (per recipe) Source
Intelligence42with thinkingOriginal model's value. Heavily compressed (~2.1 bit), intelligence is likely much lower.
- ≤ 3 bit: quantization may cost quality
- Custom kernel required
1× RTX PRO 6000
184tok/sPeak
1 request, best value reported by source, context 64k Source
Intelligenceno independent value
- ≤ 3 bit: quantization may cost quality
- Custom kernel required
1× DGX Spark
87.8tok/sPeak
1 request, synthetic prompt, with LiLiCorr/DFlash (per recipe) Source
Intelligenceno independent value
- ≤ 3 bit: quantization may cost quality
- License unclear
1× DGX Spark
75.6tok/sPeak
1 request, repetitive prompt, context 256k Source
Intelligence34with thinkingno thinking 20
- License unclear
1× DGX Spark
74.9tok/sPeak
1 request, best value reported by source Source
Intelligenceno independent value
- License unclear
1× DGX Spark
68.1tok/sPeak
1 request, synthetic speed suite (p50), with MTP (per recipe) Source
Intelligence40with thinking
- Custom kernel required
- License unclear
1× DGX Spark
52.2tok/sPeak
1 request, best value reported by source, context 3k Source
Intelligence40with thinking
- Custom kernel required
- License unclear
1× DGX Spark
35tok/sPeak
1 request, best value reported by source, context 4k Source
Intelligence40with thinking
- Experimental
- Custom kernel required
- License unclear
1× DGX Spark
30.5tok/sPeak
Best value reported by source, 1 request, prompt type not stated, greedy, 512 tokens, default serve.sh without MTP Source
Intelligenceno independent value
- ≤ 3 bit: quantization may cost quality
- Custom kernel required
1× RTX PRO 6000K2-Horizon (IFM)
30tok/sPeak
1 request, best value reported by source Source
Intelligence25with thinking
- Custom kernel required
1× DGX Spark
18.6tok/sPeak
1 request, code prompt, context 4k Source
Intelligence42with thinkingOriginal model's value. Heavily compressed (~2 bit), intelligence is likely much lower.
- ≤ 3 bit: quantization may cost quality
- Experimental
1× DGX Spark
15.7tok/sPeak
1 request, best value reported by source Source
Intelligenceno independent value
- ≤ 3 bit: quantization may cost quality
1× DGX Spark
2.8tok/sPeak
1 request, best value reported by source, context 2k Source
Intelligence40with thinkingno thinking 25
- ≤ 3 bit: quantization may cost quality
- Experimental
- Smoke test only