4× RTX PRO 6000
81.7tok/sEveryday
1 request, prose prompt, with MTP Source
Intelligence34with thinkingno thinking 22
- ≤ 3 bit: quantization may cost quality
Local AI · Creator
0xSero
32 published recipes.
Recipes
4× RTX PRO 6000
70.7tok/sEveryday
1 request, prose prompt, with MTP Source
Intelligence34with thinkingno thinking 22
- ≤ 3 bit: quantization may cost quality
4× DGX Spark
59.1tok/sEveryday
1 request, prose prompt, with DSpark (per recipe) Source
Intelligence40with thinkingno thinking 25
2× DGX Spark
~29tok/sEverydayapproximate · method not stated
1 request, prose prompt, with DSpark (7 draft tokens) (per recipe) Source
Intelligence40with thinkingno thinking 25
- ≤ 3 bit: quantization may cost quality
- Custom kernel required
1× Strix Halo
28.4tok/sEveryday
1 request, prose prompt, no speculative decoding Source
Intelligenceno independent value
- License unclear
8× RTX 3090Qwen3.5-397B-A17B (REAP 264B)
62tok/sMixed
1 request, mixed prompt set, no speculative decoding Source
Intelligenceno independent value
4× RTX PRO 6000
42.7tok/sMixed
1 request, mixed prompt set, no speculative decoding Source
Intelligenceno independent value
1× RTX 3090
33.8tok/sMixed
1 request, mean over 29 realistic document prompts (3.7–19.8K tokens), with MTP (per recipe) Source
Intelligence40with thinking
- ≤ 3 bit: quantization may cost quality
- Custom kernel required
- License unclear
4× RTX 3090Qwen3.5-35B-A3B (REAP 28B)
12.5tok/sMixed
1 request, mixed prompt set, no speculative decoding Source
Intelligenceno independent value
4× RTX PRO 6000
208tok/sPeak
1 request, best value reported by source Source
Intelligence42with thinking
4× RTX PRO 6000
202tok/sPeak
1 request, best value reported by source, context 200k Source
Intelligence40with thinkingno thinking 25
- Experimental
4× RTX 3090
137tok/sPeak
1 request, best value reported by source, context 8k Source
Intelligence34with thinkingno thinking 20
- License unclear
4× DGX Spark
118tok/sPeak
1 request, JSON prompt Source
Intelligence42with thinking
- Non-commercial
4× RTX PRO 6000
113tok/sPeak
1 request, best value reported by source Source
Intelligence29with thinking
- License unclear
2× RTX 3090
113tok/sPeak
1 request, best value reported by source, context 8k Source
Intelligence34with thinkingno thinking 20
- License unclear
4× RTX PRO 6000
87tok/sPeak
1 request, best value reported by source Source
Intelligence29with thinking
- License unclear
4× RTX PRO 6000
77tok/sPeak
1 request, best value reported by source, context 131k Source
Intelligenceno independent value
- Custom kernel required
4× RTX PRO 6000
71tok/sPeak
1 request, code prompt Source
Intelligenceno independent value
- License unclear
1× Strix Halo
70.2tok/sPeak
1 request, synthetic prompt Source
Intelligence18with thinkingno thinking 15
4× RTX 3090
50tok/sPeak
1 request, best value reported by source Source
Intelligenceno independent value
- ≤ 3 bit: quantization may cost quality
1× RTX 3090
43.2tok/sPeak
1 request, best value reported by source, context 8k Source
Intelligence34with thinkingno thinking 20
8× RTX 3090
38.5tok/sPeak
1 request, mixed prompt set Source
Intelligenceno independent value
1× DGX Spark
38.1tok/sPeak
1 request, code prompt Source
Intelligenceno independent value
- ≤ 3 bit: quantization may cost quality
- Experimental
- Custom kernel required
1× DGX Spark
24.4tok/sPeak
1 request, best value reported by source Source
Intelligenceno independent value
4× DGX Spark
22.7tok/sPeak
1 request, best value reported by source Source
Intelligence42with thinking
1× DGX Spark
18.9tok/sPeak
1 request, best value reported by source, context 4k Source
Intelligence42with thinkingOriginal model's value. Heavily compressed (~2 bit), intelligence is likely much lower.
- ≤ 3 bit: quantization may cost quality
- Custom kernel required
2× DGX Spark
17.5tok/sPeak
1 request, best value reported by source Source
Intelligence42with thinking
8× RTX 3090
13.9tok/sPeak
1 request, best value reported by source Source
Intelligence19with thinkingno thinking 15
- Experimental
1× DGX Spark
13.6tok/sPeak
1 request, best value reported by source, context 2048 Source
Intelligenceno independent value
- ≤ 3 bit: quantization may cost quality
1× DGX Spark
12.8tok/sPeak
1 request, best value reported by source, context 2048 Source
Intelligenceno independent value
- ≤ 3 bit: quantization may cost quality
1× DGX Spark
10.8tok/sPeak
1 request, best value reported by source Source
Intelligence42with thinkingOriginal model's value. Heavily compressed (~2.3 bit), intelligence is likely much lower.
- ≤ 3 bit: quantization may cost quality
1× DGX Spark
no documented speed
Intelligenceno independent value