Compare · ModelsLive · 2 picked · head to head
Gemini 3 Pro vs gpt-oss-120b
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Gemini 3 Pro wins on 15/16 benchmarks
Gemini 3 Pro wins 15 of 16 shared benchmarks. Leads in speed · agentic · arena.
Category leads
speed·Gemini 3 Proagentic·Gemini 3 Proarena·Gemini 3 Proknowledge·Gemini 3 Prolanguage·Gemini 3 Promath·gpt-oss-120breasoning·Gemini 3 Progeneral·Gemini 3 Procoding·Gemini 3 Pro
Head to head · from the data · data as of 2026-11-15
- Benchmarks · Gemini 3 Pro leads on 15 of 16 shared benchmarks (Artificial Analysis · Quality Index, APEX-Agents, Chatbot Arena Elo · Overall, Chess Puzzles, 11 more), gpt-oss-120b on 1 (HELM · Omni-MATH)
- Cheapest provider · gpt-oss-120b at AkashML and 4 other providers $0.03/M input (of 26 offers) as of 2026-11-15
- Open weights · gpt-oss-120b has open weights (on Hugging Face as openai/gpt-oss-120b)
Sources · scores from public leaderboards, list prices and provider offers from OpenRouter, price moves from the BenchGecko changelog (5% or more, held in the next daily check), Gecko tests run by BenchGecko.
Hype vs Reality
Attention vs performance
Gemini 3 Pro
#78 by perf·#9 by attention
gpt-oss-120b
#141 by perf·no signal
Vendor risk
Who is behind the model
Google DeepMind
$3.64T·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
16 benchmarks · 2 models
Gemini 3 Progpt-oss-120b
Artificial Analysis · Quality Index
Gemini 3 Pro leads by +29.7
Gemini 3 Pro
41.3
gpt-oss-120b
11.6
APEX-Agents
Gemini 3 Pro leads by +13.7
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Gemini 3 Pro
18.4
gpt-oss-120b
4.7
Chatbot Arena Elo · Overall
Gemini 3 Pro leads by +133.9
Gemini 3 Pro
1485.5
gpt-oss-120b
1351.6
Chess Puzzles
Gemini 3 Pro leads by +11.6
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Gemini 3 Pro
27.4
gpt-oss-120b
15.8
GPQA diamond
Gemini 3 Pro leads by +26.2
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Gemini 3 Pro
93.9
gpt-oss-120b
67.7
HELM · GPQA
Gemini 3 Pro leads by +11.9
Gemini 3 Pro
80.3
gpt-oss-120b
68.4
HELM · IFEval
Gemini 3 Pro leads by +4.0
Gemini 3 Pro
87.6
gpt-oss-120b
83.6
HELM · MMLU-Pro
Gemini 3 Pro leads by +10.8
Gemini 3 Pro
90.3
gpt-oss-120b
79.5
HELM · Omni-MATH
gpt-oss-120b leads by +13.2
Gemini 3 Pro
55.6
gpt-oss-120b
68.8
HELM · WildBench
Gemini 3 Pro leads by +1.4
Gemini 3 Pro
85.9
gpt-oss-120b
84.5
Metr Time Horizons
Gemini 3 Pro leads by +14.4
Gemini 3 Pro
71.0
gpt-oss-120b
56.6
OTIS Mock AIME 2024-2025
Gemini 3 Pro leads by +2.5
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Gemini 3 Pro
91.4
gpt-oss-120b
88.9
SimpleBench
Gemini 3 Pro leads by +65.2
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Gemini 3 Pro
71.7
gpt-oss-120b
6.5
SimpleQA Verified
Gemini 3 Pro leads by +59.0
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Gemini 3 Pro
72.9
gpt-oss-120b
13.9
Terminal Bench
Gemini 3 Pro leads by +50.7
Terminal-Bench 2.0 · evaluates AI agents on real terminal-based coding tasks · writing scripts, debugging, running tests, and managing projects entirely through command-line interaction. Tests both code quality and terminal fluency.
Gemini 3 Pro
69.4
gpt-oss-120b
18.7
WeirdML
Gemini 3 Pro leads by +21.8
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Gemini 3 Pro
69.9
gpt-oss-120b
48.2
Full benchmark table
| Benchmark | Gemini 3 Pro | gpt-oss-120b |
|---|---|---|
Artificial Analysis · Quality Index | 41.3 | 11.6 |
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 18.4 | 4.7 |
Chatbot Arena Elo · Overall | 1485.5 | 1351.6 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 27.4 | 15.8 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 93.9 | 67.7 |
HELM · GPQA | 80.3 | 68.4 |
HELM · IFEval | 87.6 | 83.6 |
HELM · MMLU-Pro | 90.3 | 79.5 |
HELM · Omni-MATH | 55.6 | 68.8 |
HELM · WildBench | 85.9 | 84.5 |
Metr Time Horizons | 71.0 | 56.6 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 91.4 | 88.9 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 71.7 | 6.5 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 72.9 | 13.9 |
Terminal Bench Terminal-Bench 2.0 · evaluates AI agents on real terminal-based coding tasks · writing scripts, debugging, running tests, and managing projects entirely through command-line interaction. Tests both code quality and terminal fluency. | 69.4 | 18.7 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 69.9 | 48.2 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| n/a | n/a | n/a | n/a | |
| $0.15 | $0.60 | 131K tokens (~66 books) | $2.62 |
Gemini 3 Pro vs gpt-oss-120b · questions answered from the data
Gemini 3 Pro leads on 15 of 16 shared benchmarks (Artificial Analysis · Quality Index, APEX-Agents, Chatbot Arena Elo · Overall, Chess Puzzles, 11 more), gpt-oss-120b on 1 (HELM · Omni-MATH)