PipelineScore
← Back to leaderboard
Head-to-head

Llama 3.3 70B Instruct vs openai/gpt-oss-20b

openai/gpt-oss-20b wins+13.3 pts (+17%)on the balanced composite.
meta
Llama 3.3 70B Instruct
78.7MAINLINE
local
Winner
openai/gpt-oss-20b
92.0TRUNK

Category-by-category

Higher bar wins each row.

Llama 3.3 70B Instruct
Category
openai/gpt-oss-20b
-19.380.7
Code
100.0+19.3
+4.779.7
Reason
75.0-4.7
-8.379.3
Tool Use
87.5+8.3
-22.177.9
RAG
100.0+22.1
-27.471.8
Speed
99.3+27.4
Llama 3.3 70B Instruct
Provider
meta
Released
2024-12-06
Context
128K
Lab
Community
openai/gpt-oss-20b
Provider
local
Released
Context
0K
Lab
Community