Gemini 3 Pro Preview vs GPT-OSS 120B

Detailed comparison between Gemini 3 Pro Preview and GPT-OSS 120B for RAG applications. See which LLM best meets your accuracy, performance, and cost needs.

Model Comparison

Gemini 3 Pro Preview takes the lead.

Both Gemini 3 Pro Preview and GPT-OSS 120B are powerful language models designed for RAG applications. However, their performance characteristics differ in important ways.

Why Gemini 3 Pro Preview:

  • Gemini 3 Pro Preview has 206 higher ELO rating
  • Gemini 3 Pro Preview delivers better overall quality (4.90 vs 4.85)
  • Gemini 3 Pro Preview has a 26.1% higher win rate

Overview

Key metrics

ELO Rating

Overall ranking quality

Gemini 3 Pro Preview

1522

GPT-OSS 120B

1316

Win Rate

Head-to-head performance

Gemini 3 Pro Preview

44.9%

GPT-OSS 120B

18.9%

Quality Score

Overall quality metric

Gemini 3 Pro Preview

4.90

GPT-OSS 120B

4.85

Average Latency

Response time

Gemini 3 Pro Preview

17903ms

GPT-OSS 120B

11199ms

Visual Performance Analysis

Performance

ELO Rating Comparison

Win/Loss/Tie Breakdown

Quality Across Datasets (Overall Score)

Latency Distribution (ms)

Breakdown

How the models stack up

MetricGemini 3 Pro PreviewGPT-OSS 120BDescription
Overall Performance
ELO Rating
1522
1316
Overall ranking quality based on pairwise comparisons
Win Rate
44.9%
18.9%
Percentage of comparisons won against other models
Quality Score
4.90
4.85
Average quality across all RAG metrics
Pricing & Context
Input Price per 1M
$2.00
$0.04
Cost per million input tokens
Output Price per 1M
$12.00
$0.19
Cost per million output tokens
Context Window
1049K
131K
Maximum context window size
Release Date
2025-11-18
2025-08-05
Model release date
Performance Metrics
Avg Latency
17.9s
11.2s
Average response time across all datasets

Dataset Performance

By benchmark

Comprehensive comparison of RAG quality metrics (correctness, faithfulness, grounding, relevance, completeness) and latency for each benchmark dataset.

MSMARCO

MetricGemini 3 Pro PreviewGPT-OSS 120BDescription
Quality Metrics
Correctness
4.80
4.93
Factual accuracy of responses
Faithfulness
4.80
4.90
Adherence to source material
Grounding
4.80
4.90
Citations and context usage
Relevance
5.00
4.97
Query alignment and focus
Completeness
4.87
4.87
Coverage of all aspects
Overall
4.85
4.91
Average across all metrics
Latency Metrics
Mean
13990ms
5616ms
Average response time
Min7461ms1255msFastest response time
Max26343ms20330msSlowest response time

PG

MetricGemini 3 Pro PreviewGPT-OSS 120BDescription
Quality Metrics
Correctness
4.97
4.80
Factual accuracy of responses
Faithfulness
4.97
4.80
Adherence to source material
Grounding
4.97
4.80
Citations and context usage
Relevance
5.00
4.83
Query alignment and focus
Completeness
4.80
4.73
Coverage of all aspects
Overall
4.94
4.79
Average across all metrics
Latency Metrics
Mean
25137ms
19128ms
Average response time
Min13317ms1317msFastest response time
Max62299ms69491msSlowest response time

SciFact

MetricGemini 3 Pro PreviewGPT-OSS 120BDescription
Quality Metrics
Correctness
4.93
4.87
Factual accuracy of responses
Faithfulness
4.97
4.87
Adherence to source material
Grounding
4.93
4.87
Citations and context usage
Relevance
4.93
4.80
Query alignment and focus
Completeness
4.77
4.70
Coverage of all aspects
Overall
4.91
4.82
Average across all metrics
Latency Metrics
Mean
14583ms
8854ms
Average response time
Min10135ms0msFastest response time
Max21489ms35709msSlowest response time

Explore More

Compare more LLMs

See how all LLMs stack up for RAG applications. Compare GPT-5, Claude, Gemini, and more. View comprehensive benchmarks and find the perfect LLM for your needs.