GPT OSS 20BvsMercury
GPT OSS 20B is cheaper than Mercury at $0.02/M vs $10/M input tokens.
| Spec | GPT OSS 20B | Mercury |
|---|---|---|
| Input price | $0.02/M tokens | $10/M tokens |
| Output price | $0.1/M tokens | $10/M tokens |
| Context window | 131K tokens | 32K tokens |
| Parameters | 21B (3.6B active) | Not disclosed |
| Weights | Open | — |
| Free API (OpenRouter) | No | No |
| Released | Aug 2025 | Jun 2025 |
Our Verdict
GPT OSS 20B
MercuryRunner-up
Not enough votes to call it. On the specs, GPT OSS 20B has the edge: bigger model tier, newer, bigger context window, major provider backing.
GPT OSS 20B costs 100x less per token.
Slight edge
API pricing
Cost per 1M tokens
GPT OSS 20B
Input
$0.02
500× cheaper
Output
$0.10
100× cheaper
Mercury
Input
$10.00
Output
$10.00
GPT OSS 20B is cheaper on both: 500× input, 100× output.
Where to run it
12 hosts, cheapest first
Mercury
No hosts listed on OpenRouter.
Per million tokens. Prices and uptime via OpenRouter, checked 9 Sep 2026.
Writing DNA
Style Comparison
Similarity
62%
GPT OSS 20B uses 5.4x more transitions
GPT OSS 20B
Mercury
54%Vocabulary60%
17wSentence Length31w
0.26Hedging0.78
5.9Bold4.1
3.3Lists2.8
0.10Emoji0.03
0.75Headings0.88
0.32Transitions0.06
Based on 21 + 26 text responses
Research
What we learned reading every model
FAQ
Common questions
Keep exploring
