PipelineScore
← Back to leaderboard
Head-to-head

DeepSeek R1 Distill Llama 8B vs openai/gpt-oss-20b

openai/gpt-oss-20b wins+19.8 pts (+27%)on the balanced composite.
local
Winner
openai/gpt-oss-20b
92.0TRUNK

Category-by-category

Higher bar wins each row.

DeepSeek R1 Distill Llama 8B
Category
openai/gpt-oss-20b
-29.270.8
Code
100.0+29.2
-3.371.8
Reason
75.0+3.3
-11.376.2
Tool Use
87.5+11.3
-26.773.3
RAG
100.0+26.7
-26.073.3
Speed
99.3+26.0
DeepSeek R1 Distill Llama 8B
Provider
deepseek
Released
2025-01-20
Context
128K
Lab
Community
openai/gpt-oss-20b
Provider
local
Released
Context
0K
Lab
Community