Machine

RTX 3500 Ada Generation Laptop · Core Ultra 9 185H #acaae491

93 runs of 8 models since Nov 11, 2026. Last run Nov 12, 2026.

ShareRedditX
93
runs
8
models
51
configs
18
configs run 2+ times

Hardware

GPU
NVIDIA RTX 3500 Ada Generation Laptop GPU 12 GB
CPU
Intel(R) Core(TM) Ultra 9 185H 11 cores
RAM
47 GB
OS
Ubuntu 25.10
Tools
Ollama 93
Client
0.4.99
All NVIDIA RTX 3500 Ada Generation Laptop GPU benchmarks

Models on this machine

One dot per model, at its best config. Up is better answers, right is faster. Time to finish counts every token the model wrote, thinking included, so a model that talks a lot ends up further left.

Best trade-off: quality 60+, and nothing else here is both faster and betterOther models · dot size = number of runs
below 20 tok/s: slower than you read4050607080901005102050100tok/s (log scale) →quality →outliers sit at the edges

Medians of each model's best config, 8 models.

Best config per model

sort

Runs with the same model, quant, tool, context, thinking and KV cache count as one config. Numbers are the median of its runs, and each dot is one run on the 0–100 quality scale. ★ marks the best config with at least 2 rated runs.

ornith-1.5:9b
3 runs · 3 configs
★ bestQ4_K_MOllama16k ctxthinking off
80.6
1 rated run
44.2 tok/s
gemma4:31b
5 runs · 3 configs
★ bestQ4_K_MOllama16k ctxthinking default
×2 runs
77.877.0–78.5
steady
4.6 tok/s
4.6–4.6
gemma4:12b
7 runs · 3 configs
★ bestQ4_K_MOllama8k ctxthinking default
×3 runs
75.973.1–76.4
steady
48.5 tok/s
47.7–48.9
gemma4:26b
5 runs · 3 configs
★ bestQ4_K_MOllama16k ctxthinking default
×2 runs
75.674.3–76.8
steady
40.1 tok/s
39.9–40.2
gemma4:e4b
17 runs · 13 configs
★ bestQ4_K_MOllama32k ctxthinking default
×2 runs
74.373.8–74.7
steady
99.3 tok/s
98.1–101
nemotron-3.5-bolt:e4b
7 runs · 3 configs
★ bestQ4_K_MOllama16k ctxthinking default
×2 runs
70.467.1–73.7
steady
81.9 tok/s
81.6–82.2
gemma4:e2b
36 runs · 13 configs
★ bestQ4_K_MOllama32k ctxthinking default
×7 runs
69.863.1–74.0
varies
153 tok/s
150–155
ornith:9b
13 runs · 10 configs
★ bestQ4_K_MOllama16k ctxthinking default
×3 runs
67.466.6–71.6
steady
45.2 tok/s
44.1–46.2

Compare with another machine

Run history

Nov 12, 2026
20:39ornith-1.5:9b44.7 tok/s67.220:19ornith-1.5:9b44.2 tok/s80.619:51ornith-1.5:9b36.3 tok/s56.019:08gemma4:e4b93.9 tok/s65.619:05gemma4:e4b99.4 tok/s72.019:01gemma4:e4b94.5 tok/s66.818:58gemma4:e4b101 tok/s72.918:55gemma4:e4b93.3 tok/s64.918:52gemma4:e4b99.6 tok/s76.818:48gemma4:e4b94.9 tok/s58.518:45gemma4:e4b98.2 tok/s64.518:42gemma4:e4b101 tok/s70.718:39gemma4:e4b97.8 tok/s71.318:35ornith:9b45.2 tok/s68.518:31ornith:9b47.3 tok/s69.118:26ornith:9b45.2 tok/s67.618:23ornith:9b44.4 tok/s75.618:17ornith:9b45.2 tok/s68.418:13ornith:9b45.8 tok/s75.118:08ornith:9b43.4 tok/s63.518:04ornith:9b44.3 tok/s65.318:00gemma4:e2b151 tok/s58.717:57gemma4:e2b154 tok/s67.717:54gemma4:e2b143 tok/s68.217:52gemma4:e2b151 tok/s69.317:49gemma4:e2b151 tok/s66.617:46gemma4:e2b153 tok/s70.317:43gemma4:e2b156 tok/s68.717:41gemma4:e2b150 tok/s69.417:38gemma4:e2b153 tok/s69.517:35gemma4:e2b154 tok/s71.014:42ornith:9b46.2 tok/s71.614:37gemma4:e2b155 tok/s68.214:34gemma4:e2b152 tok/s73.214:31gemma4:e2b151 tok/s73.714:28gemma4:e2b152 tok/s67.014:25gemma4:e2b150 tok/s69.814:22gemma4:e2b153 tok/s74.714:19gemma4:e2b153 tok/s64.314:16gemma4:e2b155 tok/s61.314:13gemma4:e2b152 tok/s73.414:13gemma4:31b4.5 tok/s70.813:36gemma4:e2b150 tok/s65.613:35gemma4:31b4.6 tok/s77.012:58gemma4:31b4.4 tok/s75.912:08gemma4:31b4.6 tok/s78.511:27gemma4:26b38.5 tok/s73.911:20gemma4:26b39.9 tok/s74.311:12gemma4:26b39.2 tok/s71.711:05gemma4:26b40.2 tok/s76.810:57gemma4:12b49.3 tok/s74.010:51gemma4:12b48.9 tok/s72.010:45gemma4:12b49.0 tok/s73.610:39gemma4:12b47.7 tok/s76.510:31nemotron-3.5-bolt:e4b84.3 tok/s71.510:28nemotron-3.5-bolt:e4b81.6 tok/s67.110:25nemotron-3.5-bolt:e4b85.1 tok/s64.910:21nemotron-3.5-bolt:e4b82.2 tok/s73.710:18gemma4:e4b101 tok/s74.710:15gemma4:e4b99.5 tok/s73.610:11gemma4:e4b98.1 tok/s73.810:08gemma4:e4b99.1 tok/s74.410:04ornith:9b43.6 tok/s74.109:59ornith:9b45.2 tok/s67.409:54ornith:9b42.4 tok/s60.409:38ornith:9b44.1 tok/s66.609:22gemma4:e2b154 tok/s74.009:19gemma4:e2b158 tok/s68.209:16gemma4:e2b155 tok/s63.109:13gemma4:e2b157 tok/s64.409:08gemma4:e2b×8156 tok/s69.005:54gemma4:31b4.6 tok/s78.005:09gemma4:26b40.3 tok/s76.104:08nemotron-3.5-bolt:e4b82.6 tok/s62.804:04gemma4:12b47.7 tok/s76.403:59gemma4:e4b97.6 tok/s62.303:55gemma4:e2b155 tok/s65.702:51nemotron-3.5-bolt:e4b81.5 tok/s62.102:47gemma4:12b48.9 tok/s73.102:42gemma4:e4b96.4 tok/s71.902:38gemma4:e2b155 tok/s61.400:38nemotron-3.5-bolt:e4b82.5 tok/s72.300:35gemma4:12b48.5 tok/s75.900:29gemma4:e4b97.9 tok/s71.000:25gemma4:e2b152 tok/s71.8