WWhichAITry via Vercel AI Gateway
Compare/head-to-head

Gemini Nano 2 vs GPT-OSS 20B

On-device first. Free inference, private by default. Meanwhile, gpt-oss 20b: openai's open-weights return.

Side-by-side

Specs

DimensionGemini Nano 2GPT-OSS 20B
Provider
Google
OpenAI
Released
2025-11
2025-09
Context window
32K
128K
Output max
4K
8K
Input /M
Free
$0.15
Output /M
Free
$0.60
Modalities
text
text
Open weights
no
yes
Scorecard

Dimension-by-dimension

Gemini Nano 2
GPT-OSS 20B
reasoning
5.0
7.0
coding
4.0
7.0
writing
6.0
7.0
speed
10.0
9.0
value
10.0
10.0
Verdict

Which wins, by use case

Use caseWinnerWhy
codingGPT-OSS 20Ba decisive lead on our weighted coding score
writingGPT-OSS 20Ba clear edge on our weighted writing score
chatGemini Nano 2a coin flip on our weighted chat score
agentsGPT-OSS 20Ba decisive lead on our weighted agents score
summarizationGemini Nano 2a coin flip on our weighted summarization score
translationGPT-OSS 20Ba coin flip on our weighted translation score
reasoningGPT-OSS 20Ba decisive lead on our weighted reasoning score
researchGPT-OSS 20Ba clear edge on our weighted research score
vision and multimodalGPT-OSS 20Ba clear edge on our weighted vision score
Cheapest AI models for bulk workloadsGemini Nano 2a coin flip on our weighted cheap-bulk score
Bottom line

Our take

Pick Gemini Nano 2 if you need: free on-device inference, private by design.

Pick GPT-OSS 20B if you need: single-gpu inference, open weights.

At a 500M-input / 150M-output monthly volume, Gemini Nano 2 costs roughly $0 vs GPT-OSS 20B at $165. Use our calculator to plug in your own numbers.

Keep browsing

Other comparisons