Compare/Gemma 3 1B Instruct vs gpt-oss-120B (low)

Gemma 3 1B Instructvsgpt-oss-120B (low)

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

Google

Gemma 3 1B Instruct

Input
$0/M
Output
$0/M
Speed
47 tok/s
TTFT
0.57s
OpenAI

gpt-oss-120B (low)

Input
$0.15/M
Output
$0.6/M
Speed
255 tok/s
TTFT
0.51s

Winner by Category

Cheaper
Gemma 3 1B Instruct
Faster (tok/s)
gpt-oss-120B (low)
Lower Latency
gpt-oss-120B (low)
Benchmarks (1-10)
gpt-oss-120B (low)

Pricing Comparison

MetricGemma 3 1B Instructgpt-oss-120B (low)
Input ($/M tokens)$0$0.15
Output ($/M tokens)$0$0.6
Cost for 1M input + 100K output tokens:
Gemma 3 1B Instruct$0.00
gpt-oss-120B (low)$0.21

Speed Comparison

Output Speed (tokens/s) — higher is better
Gemma 3 1B Instruct
47 tok/s
gpt-oss-120B (low)
255 tok/s
Time to First Token (seconds) — lower is better
Gemma 3 1B Instruct
0.57s
gpt-oss-120B (low)
0.51s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
5.524.5
Coding Index
0.215.5
Math Index
3.366.7
GPQA Diamond
23.7%67.2%
MMLU-Pro
13.5%77.5%
LiveCodeBench
1.7%70.7%
AIME 2025
3.3%66.7%
MATH-500
48.4%
Humanity's Last Exam
5.2%5.2%
SciCode
0.7%36.0%
IFBench
19.9%58.3%
TerminalBench
0.0%5.3%
Gemma 3 1B Instruct1 wins
10 winsgpt-oss-120B (low)

Frequently Asked Questions

Which is cheaper, Gemma 3 1B Instruct or gpt-oss-120B (low)?

Gemma 3 1B Instruct is cheaper overall. Its blended price (3:1 input/output ratio) is $0.00/M tokens vs $0.26/M for gpt-oss-120B (low).

Which model performs better on benchmarks?

gpt-oss-120B (low) wins 10 out of 12 benchmarks compared to 1 for Gemma 3 1B Instruct. See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

gpt-oss-120B (low) generates tokens faster at 255 tok/s vs 47 tok/s. However, gpt-oss-120B (low) has lower time-to-first-token (0.51s vs 0.57s).

When should I use Gemma 3 1B Instruct vs gpt-oss-120B (low)?

Choose based on your priorities: Gemma 3 1B Instruct for lower cost, gpt-oss-120B (low) for stronger benchmark performance, and gpt-oss-120B (low) for faster generation. For latency-sensitive apps, check the TTFT comparison above.