token.app › Compare › gpt-oss-safeguard-20b vs Mercury 3

gpt-oss-safeguard-20b vs Mercury 3

gpt-oss-safeguard-20b is 1.2× cheaper per token on a blended 3:1 basis. Mercury 3 accepts the larger context window (1M).

Head to head
Blended price uses a 3:1 input:output mix
gpt-oss-safeguard-20bMercury 3
ProviderOpenAIInception
Input $/1M$0.075$0.100
Output $/1M$0.300$0.350
Blended $/1M$0.131$0.163
Context window131K1M
Max output66K131K
ReleasedOct 2025Oct 2026
LicenceOpen weightsProprietary
Choose gpt-oss-safeguard-20b if…
  • Costs less per token — $0.131 vs $0.163 blended
  • Open weights — self-hostable
Choose Mercury 3 if…
  • Larger context window (1M)
  • Newer model (released Oct 2026)
FAQ
Which is better, gpt-oss-safeguard-20b or Mercury 3?
We do not hold independently-run benchmark scores covering both gpt-oss-safeguard-20b and Mercury 3, so we make no quality claim. Their pricing and specifications are compared on this page.
Is gpt-oss-safeguard-20b cheaper than Mercury 3?
gpt-oss-safeguard-20b is cheaper: $0.131 versus $0.163 per million tokens blended at 3:1 input:output. Input alone: $0.075 vs $0.100. Output alone: $0.300 vs $0.350.
What context windows do gpt-oss-safeguard-20b and Mercury 3 have?
gpt-oss-safeguard-20b: 131,072 tokens. Mercury 3: 1,000,000 tokens.
Who makes gpt-oss-safeguard-20b and Mercury 3?
gpt-oss-safeguard-20b is made by OpenAI. Mercury 3 is made by Inception.
Related comparisons
gpt-oss-safeguard-20b full details Mercury 3 full details gpt-oss-safeguard-20b vs Flint (free) gpt-oss-safeguard-20b vs MiniMax M3.1 Flash (free) gpt-oss-safeguard-20b vs DeepSeek V4.1 Flash 1017 gpt-oss-safeguard-20b vs Glimpse gpt-oss-safeguard-20b vs Xing4.0 29B A4B (free) gpt-oss-safeguard-20b vs Beam 1.0 501B A23B