Datasets:
LocalLLMAdvisor
/
local-llm-advisor-benchmarks

Dataset card Data Studio Files
xet
 Community
1
hw_id
stringclasses
13 values
device_raw
stringclasses
14 values
device_count
int64
1
1
backend
stringclasses
1 value
model
stringclasses
14 values
quant
stringclasses
4 values
params_b
float64
4.66
117
size_gib
float64
2.54
68
fa
int64
1
1
ngl
int64
99
99
threads
int64
6
128
ubatch
int64
512
512
depth
int64
0
131k
pp_n
int64
130
4.1k
pp_tps
float64
49.3
20.5k
tg_n
int64
64
768
tg_tps
float64
4.18
396
build
stringclasses
1 value
build_num
int64
11.7k
11.7k
n_samples
int64
1
3
date
stringdate
2026-10-16 00:00:00
2026-10-28 00:00:00
notes
stringlengths
250
507
engine
stringclasses
1 value
rtx-4090
NVIDIA GeForce RTX 4090
1
CUDA
Qwen3.8 27B
Q6_K
27.78
20.46
1
99
6
512
16,384
512
2,626.11
128
38.02
3e4dfa8f3
11,658
3
2026-10-26
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 580.159.03, vast.ai, CPU AMD Ryzen 5 3600 6-Core Processor, RAM 62 GiB; llama-bench model_type 'qwen35 27B Q6_K'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.8-27B-GGUF/Qwen3.8-27B-UD-Q6_K.gguf; pp stddev 76.53; tg stddev 0.07
llama.cpp
rtx-4090
NVIDIA GeForce RTX 4090
1
CUDA
Qwen3.8 27B
Q6_K
27.78
20.46
1
99
6
512
32,768
512
2,322.46
128
36.4
3e4dfa8f3
11,658
3
2026-10-26
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 580.159.03, vast.ai, CPU AMD Ryzen 5 3600 6-Core Processor, RAM 62 GiB; llama-bench model_type 'qwen35 27B Q6_K'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.8-27B-GGUF/Qwen3.8-27B-UD-Q6_K.gguf; pp stddev 46.34; tg stddev 0.06
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Gemma 4 12B
Q4_K_M
11.96
6.62
1
99
128
512
0
512
1,981.01
128
49.05
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gemma4 ?B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q4_K_M.gguf; pp stddev 84.45; tg stddev 0.10
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Gemma 4 12B
Q4_K_M
11.96
6.62
1
99
128
512
4,096
512
1,786.13
128
46.31
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gemma4 ?B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q4_K_M.gguf; pp stddev 65.16; tg stddev 0.08
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Gemma 4 12B
Q4_K_M
11.96
6.62
1
99
128
512
16,384
512
1,470.92
128
45.54
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gemma4 ?B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q4_K_M.gguf; pp stddev 59.84; tg stddev 0.08
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Gemma 4 12B
Q8_0
11.96
11.78
1
99
128
512
0
512
2,054.86
128
30.99
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gemma4 ?B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q8_0.gguf; pp stddev 83.07; tg stddev 0.04
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Gemma 4 12B
Q8_0
11.96
11.78
1
99
128
512
4,096
512
1,862.11
128
29.88
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gemma4 ?B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q8_0.gguf; pp stddev 52.65; tg stddev 0.04
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Gemma 4 12B
Q8_0
11.96
11.78
1
99
128
512
16,384
512
1,536.85
128
29.56
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gemma4 ?B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q8_0.gguf; pp stddev 32.74; tg stddev 0.03
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Gemma 4 26B-A4B
Q4_K_M
25.81
15.77
1
99
128
512
0
512
934.15
128
12.35
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gemma4 26B.A4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-26B-A4B-it-GGUF/gemma-4-26B-A4B-it-UD-Q4_K_M.gguf; pp stddev 33.75; tg s...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Gemma 4 26B-A4B
Q4_K_M
25.81
15.77
1
99
128
512
0
512
643.21
128
7.56
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gemma4 26B.A4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-26B-A4B-it-GGUF/gemma-4-26B-A4B-it-UD-Q4_K_M.gguf; pp stddev 19.37; tg s...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Gemma 4 26B-A4B
Q4_K_M
25.81
15.77
1
99
128
512
0
512
484.92
128
5.45
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gemma4 26B.A4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-26B-A4B-it-GGUF/gemma-4-26B-A4B-it-UD-Q4_K_M.gguf; pp stddev 9.32; tg st...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Gemma 4 26B-A4B
Q4_K_M
25.81
15.77
1
99
128
512
4,096
512
929.29
128
12.63
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gemma4 26B.A4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-26B-A4B-it-GGUF/gemma-4-26B-A4B-it-UD-Q4_K_M.gguf; pp stddev 15.90; tg s...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Gemma 4 26B-A4B
Q4_K_M
25.81
15.77
1
99
128
512
16,384
512
903.34
128
12.57
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gemma4 26B.A4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-26B-A4B-it-GGUF/gemma-4-26B-A4B-it-UD-Q4_K_M.gguf; pp stddev 13.33; tg s...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Mistral Small 3.2 24B
Q4_K_M
24.01
13.34
1
99
128
512
0
512
1,103.93
128
27.88
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'llama 13B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF/Mistral-Small-3.2-24B-Instruct-2506-Q4_K_M.g...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Mistral Small 3.2 24B
Q4_K_M
24.01
13.34
1
99
128
512
4,096
512
1,033.8
128
26.52
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'llama 13B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF/Mistral-Small-3.2-24B-Instruct-2506-Q4_K_M.g...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3-Coder 30B-A3B
Q4_K_M
30.53
17.28
1
99
128
512
0
512
786.26
128
10.35
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen3moe 30B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF/Qwen3-Coder-30B-A3B-Instruct-Q4_K_M.gguf; pp...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3-Coder 30B-A3B
Q4_K_M
30.53
17.28
1
99
128
512
0
512
511.19
128
5.89
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen3moe 30B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF/Qwen3-Coder-30B-A3B-Instruct-Q4_K_M.gguf; pp...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3-Coder 30B-A3B
Q4_K_M
30.53
17.28
1
99
128
512
0
512
367.39
128
4.18
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen3moe 30B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF/Qwen3-Coder-30B-A3B-Instruct-Q4_K_M.gguf; pp...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3-Coder 30B-A3B
Q4_K_M
30.53
17.28
1
99
128
512
4,096
512
758.72
128
9.57
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen3moe 30B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF/Qwen3-Coder-30B-A3B-Instruct-Q4_K_M.gguf; pp...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3-Coder 30B-A3B
Q4_K_M
30.53
17.28
1
99
128
512
16,384
512
653.44
128
9.44
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen3moe 30B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF/Qwen3-Coder-30B-A3B-Instruct-Q4_K_M.gguf; pp...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3-VL 8B
Q4_K_M
8.77
4.68
1
99
128
512
0
512
3,060.55
128
77.69
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen3vl 8B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-VL-8B-Instruct-GGUF/Qwen3-VL-8B-Instruct-Q4_K_M.gguf; pp stddev 141.63; tg std...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3-VL 8B
Q4_K_M
8.77
4.68
1
99
128
512
4,096
512
2,613.43
128
69.07
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen3vl 8B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-VL-8B-Instruct-GGUF/Qwen3-VL-8B-Instruct-Q4_K_M.gguf; pp stddev 87.17; tg stdd...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3-VL 8B
Q4_K_M
8.77
4.68
1
99
128
512
16,384
512
1,422.97
128
53.05
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen3vl 8B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-VL-8B-Instruct-GGUF/Qwen3-VL-8B-Instruct-Q4_K_M.gguf; pp stddev 44.11; tg stdd...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3-VL 8B
Q8_0
8.77
8.11
1
99
128
512
0
512
3,189.44
128
49.09
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen3vl 8B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-VL-8B-Instruct-GGUF/Qwen3-VL-8B-Instruct-Q8_0.gguf; pp stddev 142.75; tg stddev 0.08
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3-VL 8B
Q8_0
8.77
8.11
1
99
128
512
4,096
512
2,719.45
128
45.52
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen3vl 8B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-VL-8B-Instruct-GGUF/Qwen3-VL-8B-Instruct-Q8_0.gguf; pp stddev 72.12; tg stddev 0.06
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3-VL 8B
Q8_0
8.77
8.11
1
99
128
512
16,384
512
1,471.52
128
37.95
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen3vl 8B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-VL-8B-Instruct-GGUF/Qwen3-VL-8B-Instruct-Q8_0.gguf; pp stddev 22.43; tg stddev 0.04
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.5 4B
Q4_K_M
4.66
2.54
1
99
128
512
0
512
4,775.88
128
107.64
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35 4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-4B-GGUF/Qwen3.5-4B-Q4_K_M.gguf; pp stddev 365.99; tg stddev 0.21
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.5 4B
Q4_K_M
4.66
2.54
1
99
128
512
4,096
512
4,500.46
128
103.26
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35 4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-4B-GGUF/Qwen3.5-4B-Q4_K_M.gguf; pp stddev 259.86; tg stddev 0.28
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.5 4B
Q4_K_M
4.66
2.54
1
99
128
512
16,384
512
3,838.55
128
94.48
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35 4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-4B-GGUF/Qwen3.5-4B-Q4_K_M.gguf; pp stddev 177.76; tg stddev 0.24
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.5 4B
Q8_0
4.66
4.16
1
99
128
512
0
512
4,979.83
128
79.64
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35 4B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-4B-GGUF/Qwen3.5-4B-Q8_0.gguf; pp stddev 435.84; tg stddev 0.21
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.5 4B
Q8_0
4.66
4.16
1
99
128
512
4,096
512
4,717.88
128
77.02
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35 4B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-4B-GGUF/Qwen3.5-4B-Q8_0.gguf; pp stddev 228.24; tg stddev 0.13
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.5 4B
Q8_0
4.66
4.16
1
99
128
512
16,384
512
3,963.36
128
72.07
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35 4B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-4B-GGUF/Qwen3.5-4B-Q8_0.gguf; pp stddev 173.61; tg stddev 0.22
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.5 9B
Q4_K_M
9.65
5.28
1
99
128
512
0
512
2,955.05
128
70.02
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35 9B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-9B-GGUF/Qwen3.5-9B-Q4_K_M.gguf; pp stddev 137.37; tg stddev 0.09
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.5 9B
Q4_K_M
9.65
5.28
1
99
128
512
4,096
512
2,857.91
128
68.02
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35 9B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-9B-GGUF/Qwen3.5-9B-Q4_K_M.gguf; pp stddev 99.10; tg stddev 0.22
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.5 9B
Q4_K_M
9.65
5.28
1
99
128
512
16,384
512
2,581.77
128
64.14
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35 9B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-9B-GGUF/Qwen3.5-9B-Q4_K_M.gguf; pp stddev 67.85; tg stddev 0.11
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.5 9B
Q8_0
9.65
8.86
1
99
128
512
0
512
3,078.17
128
45.91
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35 9B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-9B-GGUF/Qwen3.5-9B-Q8_0.gguf; pp stddev 157.36; tg stddev 0.02
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.5 9B
Q8_0
9.65
8.86
1
99
128
512
4,096
512
2,974.84
128
45.02
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35 9B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-9B-GGUF/Qwen3.5-9B-Q8_0.gguf; pp stddev 97.93; tg stddev 0.08
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.5 9B
Q8_0
9.65
8.86
1
99
128
512
16,384
512
2,656.74
128
43.31
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35 9B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-9B-GGUF/Qwen3.5-9B-Q8_0.gguf; pp stddev 91.03; tg stddev 0.08
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.6 35B-A3B
Q4_K_M
35.95
20.6
1
99
128
512
0
512
502.05
128
7.55
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35moe 35B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.6-35B-A3B-GGUF/Qwen3.6-35B-A3B-UD-Q4_K_M.gguf; pp stddev 18.01; tg stdd...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.6 35B-A3B
Q4_K_M
35.95
20.6
1
99
128
512
0
512
369.32
128
5.22
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35moe 35B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.6-35B-A3B-GGUF/Qwen3.6-35B-A3B-UD-Q4_K_M.gguf; pp stddev 11.35; tg stdd...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.6 35B-A3B
Q4_K_M
35.95
20.6
1
99
128
512
4,096
512
481.62
128
7.45
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35moe 35B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.6-35B-A3B-GGUF/Qwen3.6-35B-A3B-UD-Q4_K_M.gguf; pp stddev 5.14; tg stdde...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
Qwen3.6 35B-A3B
Q4_K_M
35.95
20.6
1
99
128
512
16,384
512
466.14
128
6.69
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'qwen35moe 35B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.6-35B-A3B-GGUF/Qwen3.6-35B-A3B-UD-Q4_K_M.gguf; pp stddev 4.21; tg stdde...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
gpt-oss 20B
Q4_K_M
20.91
10.81
1
99
128
512
0
512
5,143.56
128
134.09
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gpt-oss 20B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gpt-oss-20b-GGUF/gpt-oss-20b-Q4_K_M.gguf; pp stddev 397.44; tg stddev 0.44
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
gpt-oss 20B
Q4_K_M
20.91
10.81
1
99
128
512
4,096
512
4,704.89
128
124.98
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gpt-oss 20B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gpt-oss-20b-GGUF/gpt-oss-20b-Q4_K_M.gguf; pp stddev 211.27; tg stddev 0.48
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
gpt-oss 20B
Q4_K_M
20.91
10.81
1
99
128
512
16,384
512
3,673.85
128
117.9
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gpt-oss 20B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gpt-oss-20b-GGUF/gpt-oss-20b-Q4_K_M.gguf; pp stddev 137.91; tg stddev 0.97
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
gpt-oss 20B
Q8_0
20.91
11.27
1
99
128
512
0
512
5,236.97
128
125.31
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gpt-oss 20B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/gpt-oss-20b-GGUF/gpt-oss-20b-Q8_0.gguf; pp stddev 384.10; tg stddev 0.52
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
gpt-oss 20B
Q8_0
20.91
11.27
1
99
128
512
4,096
512
4,746.58
128
117.31
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gpt-oss 20B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/gpt-oss-20b-GGUF/gpt-oss-20b-Q8_0.gguf; pp stddev 121.79; tg stddev 0.31
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti
1
CUDA
gpt-oss 20B
Q8_0
20.91
11.27
1
99
128
512
16,384
512
3,680.46
128
111.32
3e4dfa8f3
11,658
3
2026-10-21
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU AMD EPYC 7V12 64-Core Processor, RAM 251 GiB; llama-bench model_type 'gpt-oss 20B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/gpt-oss-20b-GGUF/gpt-oss-20b-Q8_0.gguf; pp stddev 131.77; tg stddev 0.86
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti 16GB
1
CUDA
Qwen3.6 35B-A3B
UD-Q4_K_XL
34.66
20.81
1
99
36
512
0
4,096
473.01
128
28.81
3e4dfa8f3
11,658
2
2026-10-28
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), engine llama.cpp, driver 595.84, vast.ai, CPU Intel Xeon E5-2695 v4, RAM 503.0 GiB; GGUF unsloth/Qwen3.6-35B-A3B-GGUF/Qwen3.6-35B-A3B-UD-Q4_K_XL.gguf; -fa 1 -ngl 99 -r 2; Strata comparison run; host load average 11-51 during the llama.cpp stages and 12-20 d...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti 16GB
1
CUDA
Qwen3.6 35B-A3B
UD-Q4_K_XL
34.66
20.81
1
99
36
512
4,096
4,096
405.61
128
27.27
3e4dfa8f3
11,658
2
2026-10-28
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), engine llama.cpp, driver 595.84, vast.ai, CPU Intel Xeon E5-2695 v4, RAM 503.0 GiB; GGUF unsloth/Qwen3.6-35B-A3B-GGUF/Qwen3.6-35B-A3B-UD-Q4_K_XL.gguf; -fa 1 -ngl 99 -r 2; Strata comparison run; host load average 11-51 during the llama.cpp stages and 12-20 d...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti 16GB
1
CUDA
Qwen3.6 35B-A3B
UD-Q4_K_XL
34.66
20.81
1
99
36
512
32,768
4,096
354.05
128
19.27
3e4dfa8f3
11,658
1
2026-10-28
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), engine llama.cpp, driver 595.84, vast.ai, CPU Intel Xeon E5-2695 v4, RAM 503.0 GiB; GGUF unsloth/Qwen3.6-35B-A3B-GGUF/Qwen3.6-35B-A3B-UD-Q4_K_XL.gguf; -fa 1 -ngl 99 -r 1; Strata comparison run; host load average 11-51 during the llama.cpp stages and 12-20 d...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti 16GB
1
CUDA
Qwen3.6 35B-A3B
UD-Q4_K_XL
34.66
20.81
1
99
36
512
0
130
108.47
768
31.51
3e4dfa8f3
11,658
3
2026-10-28
own run, llama-server from llama.cpp b11658 (commit 3e4dfa8f3), -fa on -ngl 99 -c 16384 --jinja, 3 realistic ~100-token chat prompts with 768-token replies (same as Strata), median of server timings; vast.ai, CPU Intel Xeon E5-2695 v4; Strata comparison run; llama-server chat; host load average 11-51 during the llama.c...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti 16GB
1
CUDA
Qwen3.6 35B-A3B
UD-Q4_K_XL
35.51
21.27
1
99
56
512
0
512
234.71
64
14.15
3e4dfa8f3
11,658
1
2026-10-28
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), engine llama.cpp, vast.ai, CPU Intel Xeon Platinum 8273CL, RAM 761.0 GiB; GGUF unsloth/Qwen3.6-35B-A3B-GGUF/Qwen3.6-35B-A3B-UD-Q4_K_XL.gguf; -fa 1 -ngl 99 -r 1; Strata comparison run; busy shared host (load average 90-150 from other tenants during the run),...
llama.cpp
rtx-5060ti-16
NVIDIA GeForce RTX 5060 Ti 16GB
1
CUDA
Qwen3.6 35B-A3B
UD-Q4_K_XL
35.51
21.27
1
99
56
512
0
130
49.33
768
11.19
3e4dfa8f3
11,658
3
2026-10-28
own run, llama-server from llama.cpp b11658 (commit 3e4dfa8f3), -fa on -ngl 99 -c 16384 --jinja, 3 realistic ~100-token chat prompts with 768-token replies (same as Strata), median of server timings; vast.ai, CPU Intel Xeon Platinum 8273CL; Strata comparison run; llama-server chat; busy shared host (load average 90-150...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Gemma 4 12B
Q4_K_M
11.96
6.62
1
99
16
512
0
512
3,167.56
128
70.13
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'gemma4 ?B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q4_K_M.gguf; pp stddev 117.31; tg st...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Gemma 4 12B
Q4_K_M
11.96
6.62
1
99
16
512
4,096
512
2,911.59
128
66.32
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'gemma4 ?B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q4_K_M.gguf; pp stddev 66.35; tg std...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Gemma 4 12B
Q4_K_M
11.96
6.62
1
99
16
512
16,384
512
2,458.89
128
65.4
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'gemma4 ?B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q4_K_M.gguf; pp stddev 45.46; tg std...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Gemma 4 26B-A4B
Q4_K_M
25.81
15.77
1
99
16
512
0
512
1,220.85
128
32.32
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'gemma4 26B.A4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-26B-A4B-it-GGUF/gemma-4-26B-A4B-it-UD-Q4_K_M.gguf; pp stdd...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Gemma 4 26B-A4B
Q4_K_M
25.81
15.77
1
99
16
512
0
512
959.86
128
26.56
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'gemma4 26B.A4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-26B-A4B-it-GGUF/gemma-4-26B-A4B-it-UD-Q4_K_M.gguf; pp stdd...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Gemma 4 26B-A4B
Q4_K_M
25.81
15.77
1
99
16
512
4,096
512
1,206.57
128
30.99
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'gemma4 26B.A4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-26B-A4B-it-GGUF/gemma-4-26B-A4B-it-UD-Q4_K_M.gguf; pp stdd...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Gemma 4 26B-A4B
Q4_K_M
25.81
15.77
1
99
16
512
16,384
512
1,140.91
128
28.58
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'gemma4 26B.A4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-26B-A4B-it-GGUF/gemma-4-26B-A4B-it-UD-Q4_K_M.gguf; pp stdd...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3-Coder 30B-A3B
Q4_K_M
30.53
17.28
1
99
16
512
0
512
924.03
128
27.67
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen3moe 30B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF/Qwen3-Coder-30B-A3B-Instruct-Q...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3-Coder 30B-A3B
Q4_K_M
30.53
17.28
1
99
16
512
0
512
702.47
128
21.85
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen3moe 30B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF/Qwen3-Coder-30B-A3B-Instruct-Q...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3-Coder 30B-A3B
Q4_K_M
30.53
17.28
1
99
16
512
4,096
512
903.69
128
26.53
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen3moe 30B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF/Qwen3-Coder-30B-A3B-Instruct-Q...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3-Coder 30B-A3B
Q4_K_M
30.53
17.28
1
99
16
512
16,384
512
783.71
128
25.07
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen3moe 30B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF/Qwen3-Coder-30B-A3B-Instruct-Q...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3-VL 8B
Q4_K_M
8.77
4.68
1
99
16
512
0
512
4,912.05
128
111.91
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen3vl 8B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-VL-8B-Instruct-GGUF/Qwen3-VL-8B-Instruct-Q4_K_M.gguf; pp stddev ...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3-VL 8B
Q4_K_M
8.77
4.68
1
99
16
512
4,096
512
4,254.17
128
100.56
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen3vl 8B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-VL-8B-Instruct-GGUF/Qwen3-VL-8B-Instruct-Q4_K_M.gguf; pp stddev ...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3-VL 8B
Q4_K_M
8.77
4.68
1
99
16
512
16,384
512
2,916.49
128
77.08
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen3vl 8B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-VL-8B-Instruct-GGUF/Qwen3-VL-8B-Instruct-Q4_K_M.gguf; pp stddev ...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3-VL 8B
Q8_0
8.77
8.11
1
99
16
512
0
512
5,040.89
128
72.39
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen3vl 8B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-VL-8B-Instruct-GGUF/Qwen3-VL-8B-Instruct-Q8_0.gguf; pp stddev 263.65; tg ...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3-VL 8B
Q8_0
8.77
8.11
1
99
16
512
4,096
512
4,357.8
128
67.5
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen3vl 8B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-VL-8B-Instruct-GGUF/Qwen3-VL-8B-Instruct-Q8_0.gguf; pp stddev 118.75; tg ...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3-VL 8B
Q8_0
8.77
8.11
1
99
16
512
16,384
512
2,971.17
128
56.26
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen3vl 8B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3-VL-8B-Instruct-GGUF/Qwen3-VL-8B-Instruct-Q8_0.gguf; pp stddev 52.86; tg s...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.5 4B
Q4_K_M
4.66
2.54
1
99
16
512
0
512
7,449.32
128
154.54
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35 4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-4B-GGUF/Qwen3.5-4B-Q4_K_M.gguf; pp stddev 601.47; tg stddev 0.0...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.5 4B
Q4_K_M
4.66
2.54
1
99
16
512
4,096
512
7,120.95
128
147.42
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35 4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-4B-GGUF/Qwen3.5-4B-Q4_K_M.gguf; pp stddev 262.51; tg stddev 0.1...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.5 4B
Q4_K_M
4.66
2.54
1
99
16
512
16,384
512
6,020.51
128
135.25
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35 4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-4B-GGUF/Qwen3.5-4B-Q4_K_M.gguf; pp stddev 228.87; tg stddev 0.1...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.5 4B
Q8_0
4.66
4.16
1
99
16
512
0
512
7,723.96
128
115.68
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35 4B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-4B-GGUF/Qwen3.5-4B-Q8_0.gguf; pp stddev 675.25; tg stddev 0.27
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.5 4B
Q8_0
4.66
4.16
1
99
16
512
4,096
512
7,406.24
128
111.81
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35 4B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-4B-GGUF/Qwen3.5-4B-Q8_0.gguf; pp stddev 289.35; tg stddev 0.26
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.5 4B
Q8_0
4.66
4.16
1
99
16
512
16,384
512
6,236.16
128
104.96
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35 4B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-4B-GGUF/Qwen3.5-4B-Q8_0.gguf; pp stddev 236.33; tg stddev 0.24
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.5 9B
Q4_K_M
9.65
5.28
1
99
16
512
0
512
4,652.29
128
101.31
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35 9B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-9B-GGUF/Qwen3.5-9B-Q4_K_M.gguf; pp stddev 256.98; tg stddev 0.1...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.5 9B
Q4_K_M
9.65
5.28
1
99
16
512
4,096
512
4,541.33
128
98.29
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35 9B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-9B-GGUF/Qwen3.5-9B-Q4_K_M.gguf; pp stddev 133.49; tg stddev 0.1...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.5 9B
Q4_K_M
9.65
5.28
1
99
16
512
16,384
512
4,073.31
128
92.76
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35 9B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-9B-GGUF/Qwen3.5-9B-Q4_K_M.gguf; pp stddev 106.97; tg stddev 0.1...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.5 9B
Q8_0
9.65
8.86
1
99
16
512
0
512
4,757.37
128
67.58
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35 9B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-9B-GGUF/Qwen3.5-9B-Q8_0.gguf; pp stddev 263.48; tg stddev 0.10
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.5 9B
Q8_0
9.65
8.86
1
99
16
512
4,096
512
4,628.9
128
66.23
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35 9B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-9B-GGUF/Qwen3.5-9B-Q8_0.gguf; pp stddev 129.85; tg stddev 0.10
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.5 9B
Q8_0
9.65
8.86
1
99
16
512
16,384
512
4,148.84
128
63.74
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35 9B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.5-9B-GGUF/Qwen3.5-9B-Q8_0.gguf; pp stddev 112.99; tg stddev 0.08
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.6 35B-A3B
Q4_K_M
35.95
20.6
1
99
16
512
0
512
706.45
128
26.53
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35moe 35B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.6-35B-A3B-GGUF/Qwen3.6-35B-A3B-UD-Q4_K_M.gguf; pp stddev ...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.6 35B-A3B
Q4_K_M
35.95
20.6
1
99
16
512
4,096
512
650.11
128
28.33
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35moe 35B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.6-35B-A3B-GGUF/Qwen3.6-35B-A3B-UD-Q4_K_M.gguf; pp stddev ...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
Qwen3.6 35B-A3B
Q4_K_M
35.95
20.6
1
99
16
512
16,384
512
635.06
128
25.26
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'qwen35moe 35B.A3B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/Qwen3.6-35B-A3B-GGUF/Qwen3.6-35B-A3B-UD-Q4_K_M.gguf; pp stddev ...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
gpt-oss 20B
Q4_K_M
20.91
10.81
1
99
16
512
0
512
2,548.26
128
48.4
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'gpt-oss 20B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gpt-oss-20b-GGUF/gpt-oss-20b-Q4_K_M.gguf; pp stddev 112.01; tg stddev...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
gpt-oss 20B
Q4_K_M
20.91
10.81
1
99
16
512
0
512
1,675.3
128
28.72
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'gpt-oss 20B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gpt-oss-20b-GGUF/gpt-oss-20b-Q4_K_M.gguf; pp stddev 87.95; tg stddev ...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
gpt-oss 20B
Q4_K_M
20.91
10.81
1
99
16
512
0
512
1,328.67
128
22.45
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'gpt-oss 20B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gpt-oss-20b-GGUF/gpt-oss-20b-Q4_K_M.gguf; pp stddev 52.89; tg stddev ...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
gpt-oss 20B
Q4_K_M
20.91
10.81
1
99
16
512
4,096
512
2,468.39
128
44.11
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'gpt-oss 20B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gpt-oss-20b-GGUF/gpt-oss-20b-Q4_K_M.gguf; pp stddev 44.95; tg stddev ...
llama.cpp
rtx-5070
NVIDIA GeForce RTX 5070
1
CUDA
gpt-oss 20B
Q4_K_M
20.91
10.81
1
99
16
512
16,384
512
2,294.97
128
43.94
3e4dfa8f3
11,658
3
2026-10-16
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.71.05, vast.ai, CPU AMD Ryzen Threadripper PRO 3955WX 16-Cores, RAM 125 GiB; llama-bench model_type 'gpt-oss 20B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gpt-oss-20b-GGUF/gpt-oss-20b-Q4_K_M.gguf; pp stddev 34.38; tg stddev ...
llama.cpp
rtx-5070ti
NVIDIA GeForce RTX 5070 Ti
1
CUDA
Gemma 4 12B
Q4_K_M
11.96
6.62
1
99
28
512
0
512
4,164.98
128
89.15
3e4dfa8f3
11,658
3
2026-10-17
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU Intel(R) Xeon(R) CPU E5-2680 v4 @ 2.40GHz, RAM 125 GiB; llama-bench model_type 'gemma4 ?B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q4_K_M.gguf; pp stddev 311.50; tg stddev...
llama.cpp
rtx-5070ti
NVIDIA GeForce RTX 5070 Ti
1
CUDA
Gemma 4 12B
Q4_K_M
11.96
6.62
1
99
28
512
4,096
512
3,737.96
128
84.14
3e4dfa8f3
11,658
3
2026-10-17
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU Intel(R) Xeon(R) CPU E5-2680 v4 @ 2.40GHz, RAM 125 GiB; llama-bench model_type 'gemma4 ?B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q4_K_M.gguf; pp stddev 230.09; tg stddev...
llama.cpp
rtx-5070ti
NVIDIA GeForce RTX 5070 Ti
1
CUDA
Gemma 4 12B
Q4_K_M
11.96
6.62
1
99
28
512
16,384
512
3,160.99
128
83.22
3e4dfa8f3
11,658
3
2026-10-17
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU Intel(R) Xeon(R) CPU E5-2680 v4 @ 2.40GHz, RAM 125 GiB; llama-bench model_type 'gemma4 ?B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q4_K_M.gguf; pp stddev 153.91; tg stddev...
llama.cpp
rtx-5070ti
NVIDIA GeForce RTX 5070 Ti
1
CUDA
Gemma 4 12B
Q8_0
11.96
11.78
1
99
28
512
0
512
4,333.81
128
58.41
3e4dfa8f3
11,658
3
2026-10-17
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU Intel(R) Xeon(R) CPU E5-2680 v4 @ 2.40GHz, RAM 125 GiB; llama-bench model_type 'gemma4 ?B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q8_0.gguf; pp stddev 325.92; tg stddev 0.18
llama.cpp
rtx-5070ti
NVIDIA GeForce RTX 5070 Ti
1
CUDA
Gemma 4 12B
Q8_0
11.96
11.78
1
99
28
512
4,096
512
3,846.63
128
56.2
3e4dfa8f3
11,658
3
2026-10-17
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU Intel(R) Xeon(R) CPU E5-2680 v4 @ 2.40GHz, RAM 125 GiB; llama-bench model_type 'gemma4 ?B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q8_0.gguf; pp stddev 202.49; tg stddev 0.18
llama.cpp
rtx-5070ti
NVIDIA GeForce RTX 5070 Ti
1
CUDA
Gemma 4 12B
Q8_0
11.96
11.78
1
99
28
512
16,384
512
3,222.2
128
55.73
3e4dfa8f3
11,658
3
2026-10-17
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU Intel(R) Xeon(R) CPU E5-2680 v4 @ 2.40GHz, RAM 125 GiB; llama-bench model_type 'gemma4 ?B Q8_0'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-12b-it-GGUF/gemma-4-12b-it-Q8_0.gguf; pp stddev 179.80; tg stddev 0.16
llama.cpp
rtx-5070ti
NVIDIA GeForce RTX 5070 Ti
1
CUDA
Gemma 4 26B-A4B
Q4_K_M
25.81
15.77
1
99
28
512
0
512
859.2
128
80.89
3e4dfa8f3
11,658
3
2026-10-17
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU Intel(R) Xeon(R) CPU E5-2680 v4 @ 2.40GHz, RAM 125 GiB; llama-bench model_type 'gemma4 26B.A4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-26B-A4B-it-GGUF/gemma-4-26B-A4B-it-UD-Q4_K_M.gguf; pp stddev 2...
llama.cpp
rtx-5070ti
NVIDIA GeForce RTX 5070 Ti
1
CUDA
Gemma 4 26B-A4B
Q4_K_M
25.81
15.77
1
99
28
512
0
512
527.13
128
59.38
3e4dfa8f3
11,658
3
2026-10-17
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU Intel(R) Xeon(R) CPU E5-2680 v4 @ 2.40GHz, RAM 125 GiB; llama-bench model_type 'gemma4 26B.A4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-26B-A4B-it-GGUF/gemma-4-26B-A4B-it-UD-Q4_K_M.gguf; pp stddev 2...
llama.cpp
rtx-5070ti
NVIDIA GeForce RTX 5070 Ti
1
CUDA
Gemma 4 26B-A4B
Q4_K_M
25.81
15.77
1
99
28
512
0
512
390.63
128
45.41
3e4dfa8f3
11,658
3
2026-10-17
own run, llama-bench from llama.cpp b11658 (commit 3e4dfa8f3), CUDA 12.8 build, driver 595.84, vast.ai, CPU Intel(R) Xeon(R) CPU E5-2680 v4 @ 2.40GHz, RAM 125 GiB; llama-bench model_type 'gemma4 26B.A4B Q4_K - Medium'; -r 3; GPU 0 only; GGUF unsloth/gemma-4-26B-A4B-it-GGUF/gemma-4-26B-A4B-it-UD-Q4_K_M.gguf; pp stddev 9...
llama.cpp