PipelineScore
← Back to leaderboard
Head-to-head

Llama 3.1 70B Instruct vs openai/gpt-oss-20b

openai/gpt-oss-20b wins+15.7 pts (+20%)on the balanced composite.
meta
Llama 3.1 70B Instruct
76.4MAINLINE
local
Winner
openai/gpt-oss-20b
92.0TRUNK

Category-by-category

Higher bar wins each row.

Llama 3.1 70B Instruct
Category
openai/gpt-oss-20b
-26.873.2
Code
100.0+26.8
+3.878.8
Reason
75.0-3.8
-10.177.4
Tool Use
87.5+10.1
-23.776.3
RAG
100.0+23.7
-22.776.5
Speed
99.3+22.7
Llama 3.1 70B Instruct
Provider
meta
Released
2024-07-23
Context
128K
Lab
Community
openai/gpt-oss-20b
Provider
local
Released
Context
0K
Lab
Community
Llama 3.1 70B Instruct vs openai/gpt-oss-20b · PipelineScore