Navigate
AI BENCHY
Advertise here

NVIDIA: Nemotron 3 Ultra vs OpenAI: gpt-oss-120b

gpt-oss-120b (medium) leads on average score with 6.1 vs 6.1. gpt-oss-120b (medium) has the lower benchmark cost at $0.019 vs $0.095. Nemotron 3 Ultra is faster at 3.87s vs 21.91s, with pass rates of 42.4% vs 50.0%.

Recommended modelNemotron 3 UltraIts score stays close to the best score here (6.1 vs 6.1), while responding about 5.7x faster than gpt-oss-120b (medium).

Last updated at: 2026-07-20

Metric Nemotron 3 Ultra Nemotron 3 Ultra none Release: 2026-06-04 Free Available gpt-oss-120b gpt-oss-120b medium Release: 2025-08-05
Score 6.1 6.1
Rank #129 #121
Reliability 10.0 10.0
Consistency 9.3 8.0
Tests Correct
Attempt pass rate 42.4% 50.0%
Flaky tests 2 5
Total Runs 66 66
Cost per result 0.000 0.221
Total Cost $0.095 $0.019
Input Price $0.600 / 1M $0.037 / 1M
Output Price $3.600 / 1M $0.170 / 1M
Total Input Tokens 101,275 108,747
Output Tokens 9,474 29,772
Reasoning Tokens 0 68,044
Response Time (avg) 3.87s 21.91s
Response Time (max) 37.50s 68.16s
Response Time (total) 85.15s 328.70s

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#129 Nemotron 3 Ultra

none
Cost
$0.000
Time
149.6s
Tokens
3,405 tok

#121 gpt-oss-120b

medium
Cost
$0.001
Time
26.7s
Tokens
555 tok

Top Models by Score

Score vs Total Cost

Response Time (avg)

Score vs Response Time (avg)

Total Output Tokens

Score vs Total Output Tokens

Category Breakdown

Coding Score Consistency Attempt pass rate Flaky tests Tests Correct Response Time (avg) Input Tokens Output Tokens Reasoning Tokens
Nemotron 3 Ultra 5.5 10.0 33.3% 0 1.02s 7,623 369 0
gpt-oss-120b 5.9 7.0 55.6% 1 38.37s 7,782 3,365 11,973

Quick Compare

Switch Comparison Pair