Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Legal Reasoning on LegalBench
Loading...
79.3
Balanced Accuracy
Llama3.1-70B
47.3616
55.6533
63.945
72.2367
Jan 20, 2026
Feb 10, 2026
Mar 4, 2026
Mar 25, 2026
Apr 16, 2026
May 7, 2026
May 29, 2026
Balanced Accuracy
Updated 2d ago
Evaluation Results
Method
Method
Links
Balanced Accuracy
Llama3.1-70B
Post-Training Stage=SFT
2026.01
79.3
Llama3.1-70B
Post-Training Stage=DPO
2026.01
79.3
SaulLM
Parameters=141B
2026.01
79.3
Llama3.1-70B
Post-Training Stage=OOB
2026.01
78
Nemotron1.5-49B
Post-Training Stage=DPO
2026.01
76.4
Nemotron1.5-49B
Post-Training Stage=SFT
2026.01
75.9
Qwen3-30B
Post-Training Stage=DPO
2026.01
75.4
Qwen3-30B
Post-Training Stage=SFT
2026.01
75.3
Nemotron1.5-49B
Post-Training Stage=OOB
2026.01
74.2
Qwen3-30B
Post-Training Stage=OOB
2026.01
74.1
TRACE-LF
Backbone=LLaMA3-8B
2026.05
59.26
TRACE-CS
Backbone=LLaMA3-8B
2026.05
59.26
Sequential LoRA
Backbone=LLaMA3-8B
2026.05
56.66
DMT
Backbone=LLaMA3-8B
2026.05
51.42
Sequential Fine-tuning
Backbone=LLaMA3-8B
2026.05
50.7
Joint Fine-tuning
Backbone=LLaMA3-8B
2026.05
48.59
Feedback
Search any
task
Search any
task