Created bymario-alka.dePowered bynoob2claw.detricoma.de
🎮
Graphics card

NVIDIA DGX SPARK (GB10)

NVIDIA128 GB VRAM
117Benchmarks
25Models
112LLM
5Harness
15 WIdle power
140 WMax power
Best generation
903,3 tok/s
gemma-4-E2B-it
∅ Generation
143,9 tok/s
Schnitt aus 112 Läufen
Best prefill
7.470 tok/s
Prompt processing
Best TTFT
105 ms
Time to First Token
Best harness rate
71%
5 Harness-Läufe
🏆
gemma-4-E2B-it läuft am schnellsten auf dieser GPU · 903,3 tok/s
Bester Prefill: 7.470 tok/s · 25 Modelle getestet

Top models

Bester gemessener Durchsatz je Modell auf NVIDIA DGX SPARK (GB10) – umschaltbar nach Generation, Prefill oder Kombiwert.

gemma-4-E2B-it5B
903,3 tok/s
Qwen3.8-Flash-Next180B
492,8 tok/s
gemma-4-E4B-it8B
488,2 tok/s
Ling-3.0-flash127.5B
253,7 tok/s
gpt-oss-20b20B
233,3 tok/s
MiniMax-M2.7230B
174,2 tok/s
Qwen3.6-35B-A3B35B
169,4 tok/s
Anzeige
Performance profile

Durchsatz auf NVIDIA DGX SPARK (GB10)

Jede Blase steht für ein Modell, eine Engine bzw. eine CPU – Position: Prompt-Verarbeitung (X) × Ausgabe (Y), Blasengröße: Anzahl Messläufe.

MODnach Modell

1.1588695792900,002.5825.1647.746Prefill (tok/s)Generation (tok/s)gemma-4-E2B-it - 903,3 tok/s Generation, 6.421 tok/s Prefill, TTFT 7.548 ms (4 Laufe)gemma-4-E2B-itQwen3.8-Flash-Next - 492,8 tok/s Generation, 1.428 tok/s Prefill, TTFT 48.245 ms (8 Laufe)Qwen3.8-Flash-Nextgemma-4-E4B-it - 488,2 tok/s Generation, 3.820 tok/s Prefill, TTFT 13.948 ms (6 Laufe)gemma-4-E4B-itNVIDIA-Nemotron-3-Nano-4B - 367,8 tok/s Generation, 2.933 tok/s Prefill, TTFT 13.303 ms (6 Laufe)NVIDIA-Nemotron-3-Nan...Ling-3.0-flash - 253,7 tok/s Generation, 1.370 tok/s Prefill, TTFT 38.536 ms (8 Laufe)Ling-3.0-flashgpt-oss-20b - 233,3 tok/s Generation, 3.963 tok/s Prefill, TTFT 2.535 ms (6 Laufe)gpt-oss-20bQwen3-Coder-30B-A3B-Instruct - 182,9 tok/s Generation, 4.998 tok/s Prefill, TTFT 2.913 ms (3 Laufe)Qwen3-Coder-30B-A3B-I...MiniMax-M2.7 - 174,2 tok/s Generation, 5.338 tok/s Prefill, TTFT 30.237 ms (2 Laufe)MiniMax-M2.7Qwen3-30B-A3B-Instruct-2507 - 169,9 tok/s Generation, 5.183 tok/s Prefill, TTFT 2.923 ms (3 Laufe)Qwen3-30B-A3B-Instruc...Qwen3.6-35B-A3B - 169,4 tok/s Generation, 3.722 tok/s Prefill, TTFT 2.640 ms (10 Laufe)Qwen3.6-35B-A3BQwen3-30B-A3B-Thinking-2507 - 161,1 tok/s Generation, 3.723 tok/s Prefill, TTFT 3.655 ms (3 Laufe)Qwen3-30B-A3B-Thinkin...Nemotron-Cascade-2-30B-A3B - 148,8 tok/s Generation, 2.940 tok/s Prefill, TTFT 4.383 ms (5 Laufe)Nemotron-Cascade-2-30...NVIDIA-Nemotron-3.5-Lightning-30B-A3B - 136,6 tok/s Generation, 2.246 tok/s Prefill, TTFT 5.100 ms (3 Laufe)NVIDIA-Nemotron-3.5-L...gpt-oss-120b - 135,1 tok/s Generation, 1.497 tok/s Prefill, TTFT 9.902 ms (2 Laufe)gpt-oss-120bMiniMax-M2.7-AWQ-4bit - 108,6 tok/s Generation, 6.325 tok/s Prefill, TTFT 7.096 ms (4 Laufe)MiniMax-M2.7-AWQ-4bitminimax-m2.5-awq - 105,6 tok/s Generation, 6.273 tok/s Prefill, TTFT 3.973 ms (2 Laufe)minimax-m2.5-awqMiniMax-M2.5 - 103,5 tok/s Generation, 6.238 tok/s Prefill, TTFT 3.967 ms (2 Laufe)MiniMax-M2.5Codestral-22B-v0.1 - 80,0 tok/s Generation, 1.200 tok/s Prefill, TTFT 150 ms (1 Lauf)Codestral-22B-v0.1NVIDIA-Nemotron-3.5-Super-VL-120B-A12B - 79,8 tok/s Generation, 3.760 tok/s Prefill, TTFT 5.878 ms (1 Lauf)NVIDIA-Nemotron-3.5-S...
gemma-4-E2B-it 903,3 tok/sQwen3.8-Flash-Next 492,8 tok/sgemma-4-E4B-it 488,2 tok/sNVIDIA-Nemotron-3-Nano-4B 367,8 tok/sLing-3.0-flash 253,7 tok/sgpt-oss-20b 233,3 tok/sQwen3-Coder-30B-A3B-Instruct 182,9 tok/sMiniMax-M2.7 174,2 tok/sQwen3-30B-A3B-Instruct-2507 169,9 tok/sQwen3.6-35B-A3B 169,4 tok/sQwen3-30B-A3B-Thinking-2507 161,1 tok/sNemotron-Cascade-2-30B-A3B 148,8 tok/sNVIDIA-Nemotron-3.5-Lightning-30B-A3B 136,6 tok/sgpt-oss-120b 135,1 tok/sMiniMax-M2.7-AWQ-4bit 108,6 tok/sminimax-m2.5-awq 105,6 tok/sMiniMax-M2.5 103,5 tok/sCodestral-22B-v0.1 80,0 tok/sNVIDIA-Nemotron-3.5-Super-VL-120B-A12B 79,8 tok/s

ENGnach Engine

1.1398555702850,002.1144.2286.342Prefill (tok/s)Generation (tok/s)llama.cpp - 903,3 tok/s Generation, 3.221 tok/s Prefill, TTFT 11.584 ms (57 Laufe)llama.cppstrata - 492,8 tok/s Generation, 912 tok/s Prefill, TTFT 85.880 ms (4 Laufe)stratavLLM - 174,2 tok/s Generation, 5.247 tok/s Prefill, TTFT 6.777 ms (18 Laufe)vLLM
llama.cpp 903,3 tok/sstrata 492,8 tok/svLLM 174,2 tok/s

CPUnach Prozessor

1.1588695792900,05531.7833.0134.243Prefill (tok/s)Generation (tok/s)NVIDIA Grace - 903,3 tok/s Generation, 3.596 tok/s Prefill, TTFT 14.431 ms (78 Laufe)NVIDIA GraceAMD Ryzen 9 9950X 16-Core Processor - 80,0 tok/s Generation, 1.200 tok/s Prefill, TTFT 150 ms (1 Lauf)AMD Ryzen 9 9950X 16-...
NVIDIA Grace 903,3 tok/sAMD Ryzen 9 9950X 16-Core Processor 80,0 tok/s
Throughput & latency

Performancebenchmark

79 veroeffentlichte Performance-Läufe auf NVIDIA DGX SPARK (GB10).

In the leaderboard →
Metric:
Size
#Model / MakerMetricsParallelGPU / CPU / RAMRuntime
1gemma-4-E2B-it5BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
903,30 tok/s TG
Prefill 6.960 · TTFT 11.382 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
2gemma-4-E2B-it5BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
860,88 tok/s TG
Prefill 6.641 · TTFT 11.413 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
3Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
492,83 tok/s TG
Prefill 1.231 · TTFT 103.873 ms
10×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMstratagodclawIQ3_XXS
4gemma-4-E4B-it8BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
488,18 tok/s TG
Prefill 4.229 · TTFT 20.915 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
5gemma-4-E4B-it8BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
484,77 tok/s TG
Prefill 4.124 · TTFT 21.088 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
6gemma-4-E2B-it5BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
478,55 tok/s TG
Prefill 6.243 · TTFT 3.601 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
7gemma-4-E2B-it5BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
461,54 tok/s TG
Prefill 5.838 · TTFT 3.795 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
8gemma-4-E4B-it8BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
455,06 tok/s TG
Prefill 3.954 · TTFT 21.269 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
9Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
420,20 tok/s TG
Prefill 656 · TTFT 131.641 ms
10×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMstratagodclawUD-IQ4_XS
10NVIDIA-Nemotron-3-Nano-4B4BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
367,75 tok/s TG
Prefill 3.397 · TTFT 29.121 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
11NVIDIA-Nemotron-3-Nano-4B4BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
354,29 tok/s TG
Prefill 3.544 · TTFT 30.045 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
12Ling-3.0-flash127.5BinclusionAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
253,67 tok/s TG
Prefill 1.397 · TTFT 48.511 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
13gemma-4-E4B-it8BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
251,39 tok/s TG
Prefill 3.708 · TTFT 6.599 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
14Ling-3.0-flash127.5BinclusionAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
250,66 tok/s TG
Prefill 1.396 · TTFT 48.986 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
15Ling-3.0-flash127.5BinclusionAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
247,76 tok/s TG
Prefill 1.399 · TTFT 49.349 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
16gemma-4-E4B-it8BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
246,68 tok/s TG
Prefill 3.165 · TTFT 7.126 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
17Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
246,45 tok/s TG
Prefill 1.171 · TTFT 47.068 ms
5×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMstratagodclawIQ3_XXS
18gemma-4-E4B-it8BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
245,62 tok/s TG
Prefill 3.740 · TTFT 6.690 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
19gpt-oss-20b20BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
233,28 tok/s TG
Prefill 4.282 · TTFT 4.702 ms
10×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
20Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
211,19 tok/s TG
Prefill 592 · TTFT 60.940 ms
5×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMstratagodclawUD-IQ4_XS
21NVIDIA-Nemotron-3-Nano-4B4BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
210,25 tok/s TG
Prefill 3.483 · TTFT 8.798 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
22NVIDIA-Nemotron-3-Nano-4B4BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
202,30 tok/s TG
Prefill 3.149 · TTFT 9.380 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
23gpt-oss-20b20BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
202,12 tok/s TG
Prefill 4.412 · TTFT 4.498 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliMXFP4
24Qwen3-Coder-30B-A3B-Instruct30BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
182,87 tok/s TG
Prefill 7.295 · TTFT 5.014 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ4_K_M
25MiniMax-M2.7230BMiniMax Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
174,20 tok/s TG
Prefill 5.042 · TTFT 47.620 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMvLLMgodclawAWQ
26Qwen3-30B-A3B-Instruct-250730BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
169,89 tok/s TG
Prefill 7.470 · TTFT 5.007 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ4_K_M
27Qwen3.6-35B-A3B35BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
169,42 tok/s TG
Prefill 2.873 · TTFT 7.817 ms
10×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
28Qwen3.6-35B-A3B35BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
167,11 tok/s TG
Prefill 2.814 · TTFT 6.937 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ4_K_M
29Qwen3-30B-A3B-Thinking-250730BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
161,05 tok/s TG
Prefill 5.256 · TTFT 6.170 ms
10×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
30gpt-oss-20b20BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
156,02 tok/s TG
Prefill 3.859 · TTFT 2.570 ms
5×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
31gpt-oss-20b20BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
148,88 tok/s TG
Prefill 4.312 · TTFT 2.296 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliMXFP4
32Nemotron-Cascade-2-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
148,79 tok/s TG
Prefill 3.362 · TTFT 6.588 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ4_K_M
33Qwen3-Coder-30B-A3B-Instruct30BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
146,91 tok/s TG
Prefill 4.736 · TTFT 2.900 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ4_K_M
34Ling-3.0-flash127.5BinclusionAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
142,76 tok/s TG
Prefill 1.244 · TTFT 17.133 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
35Ling-3.0-flash127.5BinclusionAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
140,35 tok/s TG
Prefill 1.237 · TTFT 17.315 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
36Ling-3.0-flash127.5BinclusionAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
139,82 tok/s TG
Prefill 1.244 · TTFT 17.279 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
37Qwen3-30B-A3B-Instruct-250730BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
139,24 tok/s TG
Prefill 5.091 · TTFT 2.946 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ4_K_M
38Qwen3.6-35B-A3B35BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
138,96 tok/s TG
Prefill 2.158 · TTFT 4.629 ms
5×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
39NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
136,58 tok/s TG
Prefill 2.934 · TTFT 8.734 ms
10×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
40Qwen3.6-35B-A3B35BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
135,61 tok/s TG
Prefill 2.590 · TTFT 3.731 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ4_K_M
41gpt-oss-120b120BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
135,08 tok/s TG
Prefill 1.548 · TTFT 12.948 ms
10×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
42Qwen3-30B-A3B-Thinking-250730BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
129,46 tok/s TG
Prefill 3.442 · TTFT 3.808 ms
5×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
43Nemotron-Cascade-2-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
123,54 tok/s TG
Prefill 2.897 · TTFT 3.743 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ4_K_M
44Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
122,39 tok/s TG
Prefill 3.269 · TTFT 6.004 ms
10×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMvLLMgodclawNVFP4
45NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
116,18 tok/s TG
Prefill 2.075 · TTFT 5.320 ms
5×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
46Ling-3.0-flash127.5BinclusionAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
108,98 tok/s TG
Prefill 1.446 · TTFT 66.475 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ4_K_M
47MiniMax-M2.7-AWQ-4bit230BMiniMax Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
108,63 tok/s TG
Prefill 4.568 · TTFT 13.673 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMvLLMgodclawAWQ
48MiniMax-M2.7230BMiniMax Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
106,12 tok/s TG
Prefill 5.634 · TTFT 12.854 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMvLLMgodclawAWQ
49minimax-m2.5-awq230BMiniMax Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
105,57 tok/s TG
Prefill 6.997 · TTFT 5.319 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMvLLMgodclawAWQ
50MiniMax-M2.5230BMiniMax Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
103,49 tok/s TG
Prefill 7.016 · TTFT 5.294 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMvLLMgodclawAWQ
51MiniMax-M2.7-AWQ-4bit230BMiniMax Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
103,25 tok/s TG
Prefill 6.620 · TTFT 5.066 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMvLLMgodclawAWQ
52Nemotron-Cascade-2-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
101,36 tok/s TG
Prefill 3.237 · TTFT 6.850 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ8_0
53Qwen3.6-35B-A3B35BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
101,18 tok/s TG
Prefill 6.144 · TTFT 314 ms
1×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMvLLMgodclawNVFP4
54gpt-oss-120b120BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
100,67 tok/s TG
Prefill 1.446 · TTFT 6.855 ms
5×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
55Qwen3.6-35B-A3B35BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
100,59 tok/s TG
Prefill 6.798 · TTFT 285 ms
1×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMvLLMgodclawNVFP4
56Qwen3.6-35B-A3B35BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
92,96 tok/s TG
Prefill 4.802 · TTFT 404 ms
1×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMvLLMgodclawNVFP4
57Ling-3.0-flash127.5BinclusionAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
90,05 tok/s TG
Prefill 1.600 · TTFT 43.239 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ4_K_M
58Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
88,87 tok/s TG
Prefill 1.120 · TTFT 20.453 ms
10×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawIQ3_XXS
59Qwen3-Coder-30B-A3B-Instruct30BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
86,23 tok/s TG
Prefill 2.963 · TTFT 824 ms
1×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ4_K_M
60gpt-oss-20b20BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
85,78 tok/s TG
Prefill 3.326 · TTFT 594 ms
1×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
61Nemotron-Cascade-2-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
84,61 tok/s TG
Prefill 2.849 · TTFT 3.806 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ8_0
62Qwen3-30B-A3B-Instruct-250730BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
84,18 tok/s TG
Prefill 2.987 · TTFT 816 ms
1×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ4_K_M
63Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
82,45 tok/s TG
Prefill 2.547 · TTFT 3.826 ms
5×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMvLLMgodclawNVFP4
64gpt-oss-20b20BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
82,16 tok/s TG
Prefill 3.589 · TTFT 550 ms
1×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliMXFP4
65Qwen3-30B-A3B-Thinking-250730BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
81,15 tok/s TG
Prefill 2.472 · TTFT 986 ms
1×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
66NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
81,10 tok/s TG
Prefill 1.729 · TTFT 1.247 ms
1×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
67Codestral-22B-v0.122BMistral AI Performance benchmark
80,00 tok/s TG
Prefill 1.200 · TTFT 150 ms
8×NVIDIA DGX SPARK (GB10)AMD Ryzen 9 9950X 16-Core Processor · 128 GB RAMvLLMvllm
68NVIDIA-Nemotron-3.5-Super-VL-120B-A12B120BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
79,84 tok/s TG
Prefill 3.760 · TTFT 5.878 ms
10×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMvLLMgodclawNVFP4
69MiniMax-M2.5230BMiniMax Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
77,74 tok/s TG
Prefill 5.460 · TTFT 2.641 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMvLLMgodclawAWQ
70Nemotron-Cascade-2-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
76,95 tok/s TG
Prefill 2.356 · TTFT 929 ms
1×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ4_K_M
71NVIDIA-Nemotron-3-Nano-4B4BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
76,87 tok/s TG
Prefill 2.223 · TTFT 1.006 ms
1×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
72Qwen3.6-35B-A3B35BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
76,77 tok/s TG
Prefill 1.830 · TTFT 1.055 ms
1×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
73Qwen3.6-35B-A3B35BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
74,19 tok/s TG
Prefill 4.923 · TTFT 393 ms
1×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMvLLMgodclawNVFP4
74minimax-m2.5-awq230BMiniMax Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
73,44 tok/s TG
Prefill 5.550 · TTFT 2.628 ms
5×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMvLLMgodclawAWQ
75NVIDIA-Nemotron-3-Nano-4B4BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
72,70 tok/s TG
Prefill 1.798 · TTFT 1.470 ms
1×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppgodclawQ4_K_M
76Qwen3.6-35B-A3B35BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
68,40 tok/s TG
Prefill 2.292 · TTFT 842 ms
1×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMllama.cppopenclaw_cliQ4_K_M
77MiniMax-M2.7-AWQ-4bit230BMiniMax Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
66,75 tok/s TG
Prefill 6.946 · TTFT 4.901 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMvLLMgodclawAWQ
78Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
66,02 tok/s TG
Prefill 838 · TTFT 12.155 ms
5×NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAMllama.cppgodclawIQ3_XXS
79MiniMax-M2.7-AWQ-4bit230BMiniMax Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
64,45 tok/s TG
Prefill 7.167 · TTFT 4.743 ms
10×NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMvLLMgodclawAWQ
Agent & chat rating

Harnessbenchmark

1 veroeffentlichte Harness-Läufe auf NVIDIA DGX SPARK (GB10).

In the leaderboard →
Metric:
#Model / MakerMetricsGPU / CPU / RAMRuntime
1MiniMax-M2.5230BMiniMax Harness benchmarkTool Usage V 1.0
1.245 Pkt
70,7% · 15/15 Aufg. · 301,0 s
NVIDIA DGX SPARK (GB10)NVIDIA Grace · 120 GB RAMopenclaw_cliBF16-INT4-AWQ

← All hardware

Technische Daten

Technische Daten – NVIDIA DGX SPARK (GB10)

HerstellerNVIDIA
ArchitekturGrace Blackwell
VRAM128 GB
Speicherbandbreite273 GB/s
TDP140 W
Release05/2025
Best generation903,3 tok/s
Best prefill7.470 tok/s
Models25
Messläufe (Performance)112