Share your thoughts, 1 month free Claude Pro on usSee more

Reading Comprehension on DROP

92.2F1 Score

DeepSeek-R1

Updated 5d ago

Evaluation Results

Method	Links
DeepSeek-R1 2025.01		92.2
DeepSeek-V3 2025.01		91.6
OpenAI-o1-1217 2025.01		90.2
Fine-tuned SOTA 2020.05		89.1
QDGAT 2023.05		88.4
Claude-3.5-Sonnet-1022 2025.01		88.3
Claude 3.5 Sonnet-1022 2026.03		88.3
DeepSeek-V3.2 Exp Base 2026.01		86.6
DeepSeek-V3.1 Base 2026.01		86.3
R1-Distill-Qwen-14B 2026.03		85.5
PaLM 2 2023.05		85
Llama 3 405B 2024.07		84.8
MiMo-V2-Flash Base 2026.01		84.7
MiMo-RL 7B-R-TAP 2026.03		84.5
OpenAI-o1-mini 2025.01		83.9
OpenAI o1-mini 2026.03		83.9
GPT-4o-0513 2025.01		83.7
GPT-4o 0513 2026.03		83.7
Kimi-K2 Base 2026.01		83.6
Gemini Ultra 2024.07		82.4
GPT-4 2023.05		80.9
GPT-4 2024.07		80.9
Llama 3 70B 2024.07		79.6
QuanTA 2024.05		79.4
MiMo 7B-RL 2026.03		78.7
Mixtral 8x22B 2024.07		77.5
R1-Distill-Qwen-7B 2026.03		77
LoRA 2024.05		74.3
QwQ-32B Preview 2026.03		71.2
MEDAL 2025.12		71.1
MEDAL 2025.12		71
PaLM 2023.05		70.8
QuanTA 2024.05		69
MEDAL 2025.12		65.4
Bst5 2025.12		65
Bst5 2025.12		64.3
LoRA 2024.05		61
Engram-40B 2026.01		60.7
LLaDA1.5 2025.12		60.5
Llama 2025.12		60
QuanTA 2024.05		59.6
QuanTA 2024.05		59.5
Llama 3 8B 2024.07		59.5
Full Fine-Tuning 2024.05		59.4
Parallel Adapter 2024.05		59
Engram-27B 2026.01		59
Series Adapter 2024.05		58.8
LLaDA 2025.12		58.2
Bst5 2025.12		57.7
Gemma 7B 2024.07		56.3
LoRA 2024.05		56.2
MoE-27B 2026.01		55.7
Dream 2025.12		55.4
LoRA 2024.05		54.8
LoRA 2024.05		54
27B w/ mHC 2025.12		53.9
Mistral 7B 2024.07		53
27B w/ HC 2025.12		51.6
27B Baseline 2025.12		47
Dense-4B 2026.01		41.6
GPT-3 2020.05		36.5
GPT-3 2020.05		34.3
GPT-3 2020.05		23.6
TinyLlama v1.0 2026.03		15.34
TinyLlama v1.1 2026.03		15.31
OPT-1.3B FFF 2026.03		15.29
OPT-1.3B FFF 2026.03		15.28
OPT-1.3B FF 2026.03		15.27
OPT-1.3B FFF 2026.03		15.25
OPT-1.3B 2026.03		14.32
OPT-1.3B FFF 2026.03		14.25
Pythia-1.4B 2026.03		12.27
Pythia-1.0B 2026.03		4.25