PipelineScore
← Back to leaderboard
Head-to-head

Hermes 3 Llama 3.1 70B vs DeepSeek R1 671B-A37B

DeepSeek R1 671B-A37B wins+8.2 pts (+11%)on the balanced composite.
nous
Hermes 3 Llama 3.1 70B
78.1MAINLINE
deepseek
Winner
DeepSeek R1 671B-A37B
86.3MAINLINE

Category-by-category

Higher bar wins each row.

Hermes 3 Llama 3.1 70B
Category
DeepSeek R1 671B-A37B
-6.780.8
Code
87.5+6.7
-2.080.7
Reason
82.7+2.0
-11.578.7
Tool Use
90.3+11.5
-2.880.2
RAG
82.9+2.8
-14.873.0
Speed
87.8+14.8
Hermes 3 Llama 3.1 70B
Provider
nous
Released
2024-08-16
Context
128K
Lab
Community
DeepSeek R1 671B-A37B
Provider
deepseek
Released
2025-01-20
Context
128K
Lab
Community
Hermes 3 Llama 3.1 70B vs DeepSeek R1 671B-A37B · PipelineScore