PipelineScore
← Back to leaderboard
Head-to-head

openai/gpt-oss-20b vs gpt-oss:latest

openai/gpt-oss-20b wins+0.5 pts (+1%)on the balanced composite.
local
Winner
openai/gpt-oss-20b
92.0TRUNK
local
gpt-oss:latest
91.6TRUNK

Category-by-category

Higher bar wins each row.

openai/gpt-oss-20b
Category
gpt-oss:latest
100.0
Code
100.0
-5.075.0
Reason
80.0+5.0
87.5
Tool Use
87.5
100.0
RAG
100.0
+11.299.3
Speed
88.1-11.2
openai/gpt-oss-20b
Provider
local
Released
Context
0K
Lab
Community
gpt-oss:latest
Provider
local
Released
Context
0K
Lab
Community