Skip to main content

gpt-oss-120b vs gpt-oss-20b

Specgpt-oss-120bgpt-oss-20b
Respan Index16.4 (#151)13.6 (#166)
Blended price per 1M$0.07$0.036
Context window131K131K
Input / output per 1M$0.037 / $0.17$0.018 / $0.09
Max output131K131K
Inputtexttext
Released2025-08-052025-08-05

Respan Index by category

Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.

Categorygpt-oss-120bgpt-oss-20b
Coding6.8 gpt-oss-120b4.7 gpt-oss-20b
Agents7.9 gpt-oss-120b7.8* gpt-oss-20b
Reasoning22.8 gpt-oss-120b18.5 gpt-oss-20b
Knowledge & language11.7* gpt-oss-120b7.1* gpt-oss-20b
Human preference39.0 gpt-oss-120b34.5 gpt-oss-20b

* Estimated: no published scores in that category. How the Respan Index works

Benchmarks both report

gpt-oss-120b leads on 7 of the 7 benchmarks both report. Higher is better. Each score links to where it was published.

Benchmarkgpt-oss-120bgpt-oss-20b
LMArena Elo13521318
Chess Puzzles (Epoch AI)20%4%
GPQA Diamond80.1%71.5%
OTIS Mock AIME 2024-2025 (Epoch AI)88.9%65.3%
WeirdML (Håvard Tveit Ihle)48.2%40.9%
AIME 202590%89.2%
SWE-bench Verified62.4%60.7%

FAQ

Is gpt-oss-120b better than gpt-oss-20b?

gpt-oss-120b scores 16.4 on the Respan Index (#151), ahead of gpt-oss-20b at 13.6 (#166). gpt-oss-120b leads on 7 of the 7 benchmarks both report.

Which is cheaper, gpt-oss-120b or gpt-oss-20b?

gpt-oss-120b lists at $0.037 input and $0.17 output per 1M tokens; gpt-oss-20b at $0.018 and $0.09. gpt-oss-20b is 2.0x cheaper per token at list price.

Which has the larger context window?

Both accept 131K tokens of context.

Compare other models

vs

Built for AI agents.
Break less.
Ship more.