local.aiEarly access
HardwareModelsEnergyWallReferralsContact

Search

Find people, models, hardware, or pages

Account
HardwareModelsEnergyWallReferralsContactClaim your nameAccount
All hardware
Discrete GPUNVIDIA

RTX 6000 Ada

NVIDIA RTX 6000 Ada 48GB

Used across local.ai for this visit
NVIDIA RTX 6000 Ada 48GB
Price
$8.2k
Memory
48 GB @ 960 GB/s
Power Draw
279 W ± 20 W

Price history

Mar 2025–Oct 2026

$8,219.99

CDW (US)
$7,693.99 to $8,219.99 Mar 19, 2025 to Oct 3, 2026. Dashed spans contain no price observations.
Unobserved span

Intelligence / Speed Frontier

Most Intelligent Model
local.ai intelligence index
  1. 80%
    Qwen 3.8 27B
  2. 73%
    Qwen3.6 27B
  3. 73%
    Qwen3.6 27B
  4. 70%
    Qwen3.6 35B A3B
  5. 67%
    DeepReinforce
    Ornith 1.0 35B
  6. 59%
    Gemma 4 31B It
  7. 54%
    Qwen3.6 35B A3B
  8. 53%
    Gemma 4 26B A4B It
  9. 53%
    H Company
    Holo 3.1 35B A3B
  10. 52%
    Gemma 4 26B A4B
  11. 38%
    NVIDIA Nemotron 3 Nano 30B A3B
  12. 29%
    Gemma 4 E4B It
  13. 25%
    Gemma 4 E4B It
  14. 21%
    Gemma 4 E2B It
  15. 18%
    Gemma 4 E2B It
  16. 13%
    LFM2.5 8B A1B
Fastest Model
Tasks per hour
  1. 422/h
    LFM2.5 8B A1B
  2. 364/h
    Gemma 4 E2B It
  3. 229/h
    Gemma 4 E2B It
  4. 223/h
    Gemma 4 E4B It
  5. 186/h
    Gemma 4 E4B It
  6. 74/h
    NVIDIA Nemotron 3 Nano 30B A3B
  7. 67/h
    H Company
    Holo 3.1 35B A3B
  8. 60/h
    Gemma 4 26B A4B
  9. 54/h
    Gemma 4 26B A4B It
  10. 51/h
    Qwen3.6 35B A3B
  11. 51/h
    DeepReinforce
    Ornith 1.0 35B
  12. 45/h
    Qwen3.6 35B A3B
  13. 22/h
    Gemma 4 31B It
  14. 13/h
    Qwen3.6 27B
  15. 11/h
    Qwen3.6 27B
  16. 7/h
    Qwen 3.8 27B

Models that run on RTX 6000 Ada

Qwen 3.8 27B3
Qwen 3.8 27B

Quant fp8 · Reasoning · Low

Qwen/Qwen3.8-27B-FP8
Intelligence
80%
Task Time
8m 37s
Intelligence / Joule
0.52
Disk Size
30.9GB
Power Draw
299W
Qwen 3.8 27B

Quant fp8 · Reasoning · Medium

Qwen/Qwen3.8-27B-FP8
Intelligence
80%
Task Time
8m 15s
Intelligence / Joule
0.54
Disk Size
30.9GB
Power Draw
299W
Qwen 3.8 27B

Quant fp8 · Reasoning · XHigh

Qwen/Qwen3.8-27B-FP8
Intelligence
80%
Task Time
10m 43s
Intelligence / Joule
0.44
Disk Size
30.9GB
Power Draw
287W
Qwen3.6 27B3
Qwen3.6 27B

Quant fp8

Qwen/Qwen3.6-27B-FP8
Intelligence
73%
Task Time
5m 41s
Intelligence / Joule
0.65
Disk Size
30.9GB
Power Draw
328W
Qwen3.6 27B

Quant Q4_K_M

unsloth/Qwen3.6-27B-MTP-GGUF
Intelligence
73%
Task Time
4m 31s
Intelligence / Joule
0.83
Disk Size
17.1GB
Power Draw
322W
Qwen3.6 27B

Quant nvfp4

unsloth/Qwen3.6-27B-NVFP4
Intelligence
64%
Task Time
5m 50s
Intelligence / Joule
0.60
Disk Size
23.4GB
Power Draw
308W
Qwen3.6 35B A3B17
Qwen3.6 35B A3B

Quant nvfp4

nvidia/Qwen3.6-35B-A3B-NVFP4
Intelligence
70%
Task Time
1m 19s
Intelligence / Joule
3.14
Disk Size
23.4GB
Power Draw
280W
Qwen3.6 35B A3B

Quant UD-Q4_K_M

unsloth/Qwen3.6-35B-A3B-GGUF
Intelligence
69%
Task Time
1m 48s
Intelligence / Joule
2.25
Disk Size
22.1GB
Power Draw
284W
Qwen3.6 35B A3B

Quant UD-Q4_K_S

unsloth/Qwen3.6-35B-A3B-GGUF
Intelligence
69%
Task Time
1m 47s
Intelligence / Joule
2.37
Disk Size
20.9GB
Power Draw
270W
Qwen3.6 35B A3B

Quant iq4xs

unsloth/Qwen3.6-35B-A3B-GGUF
Intelligence
68%
Task Time
2m 06s
Intelligence / Joule
2.10
Disk Size
17.7GB
Power Draw
259W
Qwen3.6 35B A3B

Quant UD-Q3_K_S

unsloth/Qwen3.6-35B-A3B-GGUF
Intelligence
67%
Task Time
1m 51s
Intelligence / Joule
2.20
Disk Size
15.4GB
Power Draw
274W
Qwen3.6 35B A3B

Quant MXFP4_MOE

unsloth/Qwen3.6-35B-A3B-GGUF
Intelligence
67%
Task Time
1m 58s
Intelligence / Joule
2.19
Disk Size
21.7GB
Power Draw
259W
Qwen3.6 35B A3B

Quant q4kxl

unsloth/Qwen3.6-35B-A3B-GGUF
Intelligence
67%
Task Time
1m 45s
Intelligence / Joule
2.50
Disk Size
22.4GB
Power Draw
254W
Qwen3.6 35B A3B

Quant q3km

unsloth/Qwen3.6-35B-A3B-GGUF
Intelligence
66%
Task Time
1m 41s
Intelligence / Joule
2.44
Disk Size
16.6GB
Power Draw
269W
Qwen3.6 35B A3B

Quant UD-Q3_K_XL

unsloth/Qwen3.6-35B-A3B-GGUF
Intelligence
66%
Task Time
1m 39s
Intelligence / Joule
2.34
Disk Size
16.8GB
Power Draw
286W
Qwen3.6 35B A3B

Quant iq3xxs

unsloth/Qwen3.6-35B-A3B-GGUF
Intelligence
66%
Task Time
2m 18s
Intelligence / Joule
1.76
Disk Size
13.2GB
Power Draw
272W
Qwen3.6 35B A3B

Quant awq-4bit

cyankiwi/Qwen3.6-35B-A3B-AWQ-4bit
Intelligence
65%
Task Time
1m 53s
Intelligence / Joule
2.75
Disk Size
25.0GB
Power Draw
209W
Qwen3.6 35B A3B

Quant UD-Q2_K_XL

unsloth/Qwen3.6-35B-A3B-GGUF
Intelligence
64%
Task Time
1m 58s
Intelligence / Joule
1.98
Disk Size
12.3GB
Power Draw
276W
Qwen3.6 35B A3B

Quant UD-IQ3_S

unsloth/Qwen3.6-35B-A3B-GGUF
Intelligence
64%
Task Time
1m 58s
Intelligence / Joule
1.95
Disk Size
13.7GB
Power Draw
278W
Qwen3.6 35B A3B

Quant UD-IQ2_M

unsloth/Qwen3.6-35B-A3B-GGUF
Intelligence
61%
Task Time
2m 01s
Intelligence / Joule
1.94
Disk Size
11.5GB
Power Draw
262W
Qwen3.6 35B A3B

Quant iq2xxs

unsloth/Qwen3.6-35B-A3B-GGUF
Intelligence
61%
Task Time
1m 55s
Intelligence / Joule
2.02
Disk Size
10.8GB
Power Draw
260W
Qwen3.6 35B A3B

Quant nvfp4

unsloth/Qwen3.6-35B-A3B-NVFP4
Intelligence
59%
Task Time
1m 36s
Intelligence / Joule
2.61
Disk Size
26.5GB
Power Draw
235W
Qwen3.6 35B A3B Fast

Quant nvfp4

unsloth/Qwen3.6-35B-A3B-NVFP4-Fast
Intelligence
59%
Task Time
1m 43s
Intelligence / Joule
2.29
Disk Size
23.6GB
Power Draw
249W
DeepReinforceOrnith 1.0 35B3
Ornith 1.0 35B

Quant Q4_K_M

DeepReinforcedeepreinforce-ai/Ornith-1.0-35B-GGUF
Intelligence
67%
Task Time
1m 10s
Intelligence / Joule
3.47
Disk Size
21.2GB
Power Draw
277W
Ornith 1.0 35B

Quant fp8

DeepReinforcedeepreinforce-ai/Ornith-1.0-35B-FP8
Intelligence
66%
Task Time
1m 42s
Intelligence / Joule
2.19
Disk Size
37.7GB
Power Draw
296W
Ornith 1.0 35B

Quant nvfp4

sakamakismile/Ornith-1.0-35B-NVFP4
Intelligence
53%
Task Time
1m 21s
Intelligence / Joule
2.36
Disk Size
21.9GB
Power Draw
279W
Gemma 4 31B It3
Gemma 4 31B It

Quant Q4_K_M

unsloth/gemma-4-31B-it-GGUF
Intelligence
59%
Task Time
2m 46s
Intelligence / Joule
1.07
Disk Size
18.3GB
Power Draw
335W
Gemma 4 31B It

Quant fp8

RedHatAI/gemma-4-31B-it-FP8-block
Intelligence
56%
Task Time
6m 14s
Intelligence / Joule
0.47
Disk Size
33.3GB
Power Draw
320W
Gemma 4 31B It

Quant awq-4bit

cyankiwi/gemma-4-31B-it-AWQ-4bit
Intelligence
54%
Task Time
1m 17s
Intelligence / Joule
3.19
Disk Size
20.9GB
Power Draw
223W
InternScienceAgents A12
Agents A1

Quant q4km

InternScienceInternScience/Agents-A1-Q4_K_M-GGUF
Intelligence
57%
Task Time
1m 23s
Intelligence / Joule
2.66
Disk Size
21.2GB
Power Draw
259W
Agents A1

Quant fp8

InternScienceInternScience/Agents-A1-FP8
Intelligence
55%
Task Time
1m 47s
Intelligence / Joule
1.81
Disk Size
37.7GB
Power Draw
282W
Gemma 4 26B A4B3
Gemma 4 26B A4B It

Quant fp8

RedHatAI/gemma-4-26B-A4B-it-FP8-dynamic
Intelligence
54%
Task Time
1m 23s
Intelligence / Joule
2.56
Disk Size
28.6GB
Power Draw
253W
Gemma 4 26B A4B It

Quant awq-4bit

cyankiwi/gemma-4-26B-A4B-it-AWQ-4bit
Intelligence
53%
Task Time
1m 06s
Intelligence / Joule
3.76
Disk Size
17.2GB
Power Draw
214W
Gemma 4 26B A4B

Quant nvfp4

nvidia/Gemma-4-26B-A4B-NVFP4
Intelligence
52%
Task Time
1m
Intelligence / Joule
4.57
Disk Size
18.8GB
Power Draw
190W
Bonsai 27B2
Ternary Bonsai 27B

Quant awq-4bit

prism-ml/Ternary-Bonsai-27B-AWQ-4bit
Intelligence
53%
Task Time
9m 03s
Intelligence / Joule
0.34
Disk Size
18.7GB
Power Draw
284W
Bonsai 27B

Quant awq-4bit

prism-ml/Bonsai-27B-AWQ-4bit
Intelligence
41%
Task Time
10m 22s
Intelligence / Joule
0.22
Disk Size
18.7GB
Power Draw
299W
H CompanyHolo 3.1 35B A3B2
Holo 3.1 35B A3B

Quant nvfp4

H CompanyHcompany/Holo-3.1-35B-A3B-NVFP4
Intelligence
53%
Task Time
54s
Intelligence / Joule
3.94
Disk Size
23.7GB
Power Draw
250W
Holo 3.1 35B A3B

Quant q4_k_m

H CompanyHcompany/Holo-3.1-35B-A3B-GGUF
Intelligence
52%
Task Time
57s
Intelligence / Joule
3.18
Disk Size
21.3GB
Power Draw
288W
Gemma 4 12B It3
Gemma 4 12B It

Quant q4km

unsloth/gemma-4-12b-it-GGUF
Intelligence
52%
Task Time
1m 50s
Intelligence / Joule
1.59
Disk Size
7.1GB
Power Draw
297W
Gemma 4 12B It

Quant —

google/gemma-4-12B-it
Intelligence
45%
Task Time
2m 18s
Intelligence / Joule
1.50
Disk Size
23.9GB
Power Draw
219W
Gemma 4 12B It Qat W4a16 Ct

Quant —

google/gemma-4-12B-it-qat-w4a16-ct
Intelligence
42%
Task Time
1m 55s
Intelligence / Joule
1.86
Disk Size
10.3GB
Power Draw
197W
Qwen3.5 9B Base3
Qwen3.5 9B

Quant Q4_K_M

unsloth/Qwen3.5-9B-GGUF
Intelligence
51%
Task Time
2m 05s
Intelligence / Joule
1.23
Disk Size
5.7GB
Power Draw
331W
Qwen3.5 9B

Quant —

Qwen/Qwen3.5-9B
Intelligence
30%
Task Time
3m 50s
Intelligence / Joule
0.56
Disk Size
19.3GB
Power Draw
238W
Qwen3.5 9B

Quant awq-4bit

cyankiwi/Qwen3.5-9B-AWQ-4bit
Intelligence
27%
Task Time
2m 41s
Intelligence / Joule
0.74
Disk Size
9.1GB
Power Draw
223W
Qwen3.5 9B1
Qwen3.5 9B

Quant fp8

lovedheart/Qwen3.5-9B-FP8
Intelligence
46%
Task Time
4m 28s
Intelligence / Joule
0.58
Disk Size
12.2GB
Power Draw
297W
Gpt Oss 20B1
Gpt Oss 20B

Quant mxfp4

openai/gpt-oss-20b
Intelligence
42%
Task Time
1m 03s
Intelligence / Joule
2.29
Disk Size
13.8GB
Power Draw
293W
Nemotron 3 Nano 30B A3B2
NVIDIA Nemotron 3 Nano 30B A3B

Quant fp8

nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-FP8
Intelligence
38%
Task Time
59s
Intelligence / Joule
2.22
Disk Size
32.7GB
Power Draw
290W
NVIDIA Nemotron 3 Nano 30B A3B

Quant nvfp4

nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-NVFP4
Intelligence
38%
Task Time
49s
Intelligence / Joule
0.79
Disk Size
19.3GB
Power Draw
977W
Nemotron 3 Nano Omni 30B A3B Reasoning1
Nemotron 3 Nano Omni 30B A3B Reasoning

Quant nvfp4

nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-NVFP4
Intelligence
33%
Task Time
49s
Intelligence / Joule
2.34
Disk Size
22.4GB
Power Draw
286W
Gemma 4 E4B It3
Gemma 4 E4B It

Quant Q4_K_M

unsloth/gemma-4-E4B-it-GGUF
Intelligence
29%
Task Time
19s
Intelligence / Joule
5.12
Disk Size
5.0GB
Power Draw
290W
Gemma 4 E4B It

Quant fp8

PprithivMLmods/gemma-4-E4B-it-FP8
Intelligence
27%
Task Time
32s
Intelligence / Joule
3.28
Disk Size
13.3GB
Power Draw
257W
Gemma 4 E4B It

Quant fp8

leon-se/gemma-4-E4B-it-FP8-Dynamic
Intelligence
25%
Task Time
16s
Intelligence / Joule
8.90
Disk Size
13.3GB
Power Draw
174W
Gemma 4 E2B It3
Gemma 4 E2B It

Quant UD-Q8_K_XL

unsloth/gemma-4-E2B-it-GGUF
Intelligence
21%
Task Time
16s
Intelligence / Joule
5.21
Disk Size
5.3GB
Power Draw
260W
Gemma 4 E2B It

Quant fp8

leon-se/gemma-4-E2B-it-FP8-Dynamic
Intelligence
18%
Task Time
10s
Intelligence / Joule
12.29
Disk Size
9.2GB
Power Draw
147W
Gemma 4 E2B It

Quant UD-Q4_K_XL

unsloth/gemma-4-E2B-it-GGUF
Intelligence
18%
Task Time
12s
Intelligence / Joule
6.30
Disk Size
3.2GB
Power Draw
244W
LFM2.5 8B A1B2
LFM2.5 8B A1B

Quant Q8_0

LiquidAI/LFM2.5-8B-A1B-GGUF
Intelligence
13%
Task Time
9s
Intelligence / Joule
5.75
Disk Size
9.0GB
Power Draw
268W
LFM2.5 8B A1B

Quant —

LiquidAI/LFM2.5-8B-A1B
Intelligence
10%
Task Time
9s
Intelligence / Joule
3.74
Disk Size
16.9GB
Power Draw
277W
Qwen 3.8 27B3 variations—80%0.548m 15s31GB287W
Qwen 3.8 27BQwen/Qwen3.8-27B-FP8fp8 · Reasoning · Low80%0.528m 37s31GB299W
Qwen 3.8 27BQwen/Qwen3.8-27B-FP8fp8 · Reasoning · Medium80%0.548m 15s31GB299W
Qwen 3.8 27BQwen/Qwen3.8-27B-FP8fp8 · Reasoning · XHigh80%0.4410m 43s31GB287W
Qwen3.6 27B3 variations—73%0.834m 31s17GB308W
Qwen3.6 27BQwen/Qwen3.6-27B-FP8fp873%0.655m 41s31GB328W
Qwen3.6 27Bunsloth/Qwen3.6-27B-MTP-GGUFQ4_K_M73%0.834m 31s17GB322W
Qwen3.6 27Bunsloth/Qwen3.6-27B-NVFP4nvfp464%0.605m 50s23GB308W
Qwen3.6 35B A3B17 variations—70%3.141m 19s11GB209W
Qwen3.6 35B A3Bnvidia/Qwen3.6-35B-A3B-NVFP4nvfp470%3.141m 19s23GB280W
Qwen3.6 35B A3Bunsloth/Qwen3.6-35B-A3B-GGUFUD-Q4_K_M69%2.251m 48s22GB284W
Qwen3.6 35B A3Bunsloth/Qwen3.6-35B-A3B-GGUFUD-Q4_K_S69%2.371m 47s21GB270W
Qwen3.6 35B A3Bunsloth/Qwen3.6-35B-A3B-GGUFiq4xs68%2.102m 06s18GB259W
Qwen3.6 35B A3Bunsloth/Qwen3.6-35B-A3B-GGUFUD-Q3_K_S67%2.201m 51s15GB274W
Qwen3.6 35B A3Bunsloth/Qwen3.6-35B-A3B-GGUFMXFP4_MOE67%2.191m 58s22GB259W
Qwen3.6 35B A3Bunsloth/Qwen3.6-35B-A3B-GGUFq4kxl67%2.501m 45s22GB254W
Qwen3.6 35B A3Bunsloth/Qwen3.6-35B-A3B-GGUFq3km66%2.441m 41s17GB269W
Qwen3.6 35B A3Bunsloth/Qwen3.6-35B-A3B-GGUFUD-Q3_K_XL66%2.341m 39s17GB286W
Qwen3.6 35B A3Bunsloth/Qwen3.6-35B-A3B-GGUFiq3xxs66%1.762m 18s13GB272W
Qwen3.6 35B A3Bcyankiwi/Qwen3.6-35B-A3B-AWQ-4bitawq-4bit65%2.751m 53s25GB209W
Qwen3.6 35B A3Bunsloth/Qwen3.6-35B-A3B-GGUFUD-Q2_K_XL64%1.981m 58s12GB276W
Qwen3.6 35B A3Bunsloth/Qwen3.6-35B-A3B-GGUFUD-IQ3_S64%1.951m 58s14GB278W
Qwen3.6 35B A3Bunsloth/Qwen3.6-35B-A3B-GGUFUD-IQ2_M61%1.942m 01s12GB262W
Qwen3.6 35B A3Bunsloth/Qwen3.6-35B-A3B-GGUFiq2xxs61%2.021m 55s11GB260W
Qwen3.6 35B A3Bunsloth/Qwen3.6-35B-A3B-NVFP4nvfp459%2.611m 36s26GB235W
Qwen3.6 35B A3B Fastunsloth/Qwen3.6-35B-A3B-NVFP4-Fastnvfp459%2.291m 43s24GB249W
DeepReinforceOrnith 1.0 35B3 variations—67%3.471m 10s21GB277W
DeepReinforceOrnith 1.0 35BDeepReinforcedeepreinforce-ai/Ornith-1.0-35B-GGUFQ4_K_M67%3.471m 10s21GB277W
DeepReinforceOrnith 1.0 35BDeepReinforcedeepreinforce-ai/Ornith-1.0-35B-FP8fp866%2.191m 42s38GB296W
DeepReinforceOrnith 1.0 35Bsakamakismile/Ornith-1.0-35B-NVFP4nvfp453%2.361m 21s22GB279W
Gemma 4 31B It3 variations—59%3.191m 17s18GB223W
Gemma 4 31B Itunsloth/gemma-4-31B-it-GGUFQ4_K_M59%1.072m 46s18GB335W
Gemma 4 31B ItRedHatAI/gemma-4-31B-it-FP8-blockfp856%0.476m 14s33GB320W
Gemma 4 31B Itcyankiwi/gemma-4-31B-it-AWQ-4bitawq-4bit54%3.191m 17s21GB223W
InternScienceAgents A12 variations—57%2.661m 23s21GB259W
InternScienceAgents A1InternScienceInternScience/Agents-A1-Q4_K_M-GGUFq4km57%2.661m 23s21GB259W
InternScienceAgents A1InternScienceInternScience/Agents-A1-FP8fp855%1.811m 47s38GB282W
Gemma 4 26B A4B3 variations—54%4.571m17GB190W
Gemma 4 26B A4B ItRedHatAI/gemma-4-26B-A4B-it-FP8-dynamicfp854%2.561m 23s29GB253W
Gemma 4 26B A4B Itcyankiwi/gemma-4-26B-A4B-it-AWQ-4bitawq-4bit53%3.761m 06s17GB214W
Gemma 4 26B A4Bnvidia/Gemma-4-26B-A4B-NVFP4nvfp452%4.571m19GB190W
Bonsai 27B2 variations—53%0.349m 03s19GB284W
Ternary Bonsai 27Bprism-ml/Ternary-Bonsai-27B-AWQ-4bitawq-4bit53%0.349m 03s19GB284W
Bonsai 27Bprism-ml/Bonsai-27B-AWQ-4bitawq-4bit41%0.2210m 22s19GB299W
H CompanyHolo 3.1 35B A3B2 variations—53%3.9454s21GB250W
H CompanyHolo 3.1 35B A3BH CompanyHcompany/Holo-3.1-35B-A3B-NVFP4nvfp453%3.9454s24GB250W
H CompanyHolo 3.1 35B A3BH CompanyHcompany/Holo-3.1-35B-A3B-GGUFq4_k_m52%3.1857s21GB288W
Gemma 4 12B It3 variations—52%1.861m 50s7.1GB197W
Gemma 4 12B Itunsloth/gemma-4-12b-it-GGUFq4km52%1.591m 50s7.1GB297W
Gemma 4 12B Itgoogle/gemma-4-12B-it—45%1.502m 18s24GB219W
Gemma 4 12B It Qat W4a16 Ctgoogle/gemma-4-12B-it-qat-w4a16-ct—42%1.861m 55s10GB197W
Qwen3.5 9B Base3 variations—51%1.232m 05s5.7GB223W
Qwen3.5 9Bunsloth/Qwen3.5-9B-GGUFQ4_K_M51%1.232m 05s5.7GB331W
Qwen3.5 9BQwen/Qwen3.5-9B—30%0.563m 50s19GB238W
Qwen3.5 9Bcyankiwi/Qwen3.5-9B-AWQ-4bitawq-4bit27%0.742m 41s9.1GB223W
Qwen3.5 9B1 variationfp846%0.584m 28s12GB297W
Qwen3.5 9Blovedheart/Qwen3.5-9B-FP8fp846%0.584m 28s12GB297W
Gpt Oss 20B1 variationmxfp442%2.291m 03s14GB293W
Gpt Oss 20Bopenai/gpt-oss-20bmxfp442%2.291m 03s14GB293W
Nemotron 3 Nano 30B A3B2 variations—38%2.2249s19GB290W
NVIDIA Nemotron 3 Nano 30B A3Bnvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-FP8fp838%2.2259s33GB290W
NVIDIA Nemotron 3 Nano 30B A3Bnvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-NVFP4nvfp438%0.7949s19GB977W
Nemotron 3 Nano Omni 30B A3B Reasoning1 variationnvfp433%2.3449s22GB286W
Nemotron 3 Nano Omni 30B A3B Reasoningnvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-NVFP4nvfp433%2.3449s22GB286W
Gemma 4 E4B It3 variations—29%8.9016s5.0GB174W
Gemma 4 E4B Itunsloth/gemma-4-E4B-it-GGUFQ4_K_M29%5.1219s5.0GB290W
Gemma 4 E4B ItPprithivMLmods/gemma-4-E4B-it-FP8fp827%3.2832s13GB257W
Gemma 4 E4B Itleon-se/gemma-4-E4B-it-FP8-Dynamicfp825%8.9016s13GB174W
Gemma 4 E2B It3 variations—21%12.2910s3.2GB147W
Gemma 4 E2B Itunsloth/gemma-4-E2B-it-GGUFUD-Q8_K_XL21%5.2116s5.3GB260W
Gemma 4 E2B Itleon-se/gemma-4-E2B-it-FP8-Dynamicfp818%12.2910s9.2GB147W
Gemma 4 E2B Itunsloth/gemma-4-E2B-it-GGUFUD-Q4_K_XL18%6.3012s3.2GB244W
LFM2.5 8B A1B2 variations—13%5.759s9.0GB268W
LFM2.5 8B A1BLiquidAI/LFM2.5-8B-A1B-GGUFQ8_013%5.759s9.0GB268W
LFM2.5 8B A1BLiquidAI/LFM2.5-8B-A1B—10%3.749s17GB277W
local.aiIndependent local-AI benchmarksLive benchmark data
ContactAccountReferralsAbout