Compare/gpt-oss-120b (high) vs Gemini 3 Pro Preview (high)

gpt-oss-120b (high)vsGemini 3 Pro Preview (high)

Side-by-side comparison of pricing, 12 benchmarks, and generation speed.

OpenAI

gpt-oss-120b (high)

Input
$0.15/M
Output
$0.6/M
Speed
217 tok/s
TTFT
0.54s
Google

Gemini 3 Pro Preview (high)

Input
$2/M
Output
$12/M
Speed
TTFT

Winner by Category

Cheaper
gpt-oss-120b (high)
Faster (tok/s)
gpt-oss-120b (high)
Lower Latency
Gemini 3 Pro Preview (high)
Benchmarks (1-10)
Gemini 3 Pro Preview (high)

Pricing Comparison

Metricgpt-oss-120b (high)Gemini 3 Pro Preview (high)
Input ($/M tokens)$0.15$2
Output ($/M tokens)$0.6$12
Cost for 1M input + 100K output tokens:
gpt-oss-120b (high)$0.21
Gemini 3 Pro Preview (high)$3.20

Speed Comparison

Output Speed (tokens/s) — higher is better
gpt-oss-120b (high)
217 tok/s
Gemini 3 Pro Preview (high)
Time to First Token (seconds) — lower is better
gpt-oss-120b (high)
0.54s
Gemini 3 Pro Preview (high)

Editorial Analysis

Verdict. Gemini 3 Pro Preview (high) takes the aggregate benchmark matchup 10–1 across 11 categories. Real workloads usually care about a handful of specific tasks — see the per-benchmark table above.

Pricing. Both models sit in the budget / mid-tier bracket for output-token pricing. At 0.0× the per-million-token cost, gpt-oss-120b (high) is meaningfully cheaper if your traffic is output-heavy (long completions, document generation, agent loops). gpt-oss-120b (high) makes more sense when output volume is low and absolute reasoning quality justifies the premium.

Strengths. gpt-oss-120b (high) is strongest on AIME 2025 (93%), Math Index (93.4), LiveCodeBench (88%). Gemini 3 Pro Preview (high) leads on Math Index (95.7), AIME 2025 (96%), LiveCodeBench (92%).

Speed. On throughput, gpt-oss-120b (high) generates tokens at 217 tok/s versus 0 tok/s — about 100% faster. On time-to-first-token, Gemini 3 Pro Preview (high) responds in 0ms vs 535ms, which matters most for chat-style UIs.

Provider. OpenAI and Google sell to overlapping but distinct developer audiences: OpenAI tends to ship frontier reasoning models with premium positioning, while Google often prices more aggressively. Your existing vendor relationships, billing, and SLA preferences may matter as much as the raw numbers above.

Workload cost. Workload scenarios (per million requests at 30M input + 15M output tokens): gpt-oss-120b (high) costs $13.50 ($162/year); Gemini 3 Pro Preview (high) costs $240.00 ($2880/year). At a smaller 5M-input/2M-output scale (single-developer tool or prototype): gpt-oss-120b (high) ≈ $1.95/run, Gemini 3 Pro Preview (high) ≈ $34.00/run. At agent/realtime scale (200M input / 100M output per million requests): gpt-oss-120b (high) ≈ $90/run, Gemini 3 Pro Preview (high) ≈ $1600/run. gpt-oss-120b (high) becomes more attractive at higher volume — the absolute per-token pricing difference compounds when you ship at scale.

Recommendation. Both models have legitimate use cases — the right answer depends on whether you are optimizing for benchmark ceiling, latency, or unit cost. Start with the cheaper / faster model, evaluate against your specific task, and only switch if the upgrade shows a meaningful lift.

Head-to-head deltas

  • Gemini 3 Pro Preview (high) wins 9 more benchmarks than its opponent — a margin wide enough to call the comparison settled on benchmark terms alone.
  • On throughput, gpt-oss-120b (high) is 21658.20× faster (217 tok/s vs 0 tok/s). For streaming chat or real-time agents this alone often flips the recommendation.
  • Time-to-first-token differs by 5350.0× — Gemini 3 Pro Preview (high) responds in 0ms vs 535ms. For interactive chat UIs this can matter more than raw benchmark wins.
  • gpt-oss-120b (high) is 20.00× cheaper per million output tokens than Gemini 3 Pro Preview (high). At scale, that price ratio is roughly your monthly bill — the deciding factor for most production teams.
  • Aggregate benchmark score (sum across 12 categories, capped at 100): gpt-oss-120b (high) = 153, Gemini 3 Pro Preview (high) = 141. Within 15% — effectively equivalent if both meet the threshold your product requires.

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
23.839.6
Coding Index
30.4
Math Index
93.495.7
GPQA Diamond
78.2%90.8%
MMLU-Pro
80.8%89.8%
LiveCodeBench
87.8%91.7%
AIME 2025
93.4%95.7%
MATH-500
Humanity's Last Exam
18.5%37.2%
SciCode
38.9%56.1%
IFBench
69.0%70.4%
TerminalBench
23.5%41.7%
gpt-oss-120b (high)1 wins
10 winsGemini 3 Pro Preview (high)

Frequently Asked Questions

Which is cheaper, gpt-oss-120b (high) or Gemini 3 Pro Preview (high)?

gpt-oss-120b (high) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.26/M tokens vs $4.50/M for Gemini 3 Pro Preview (high).

Which model performs better on benchmarks?

Gemini 3 Pro Preview (high) wins 10 out of 12 benchmarks compared to 1 for gpt-oss-120b (high). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

gpt-oss-120b (high) generates tokens faster at 217 tok/s vs 0 tok/s. However, Gemini 3 Pro Preview (high) has lower time-to-first-token (0.00s vs 0.54s).

When should I use gpt-oss-120b (high) vs Gemini 3 Pro Preview (high)?

Choose based on your priorities: gpt-oss-120b (high) for lower cost, Gemini 3 Pro Preview (high) for stronger benchmark performance, and gpt-oss-120b (high) for faster generation. For latency-sensitive apps, check the TTFT comparison above.