Qwen2 Instruct 72B
AlibabaQwenOpen Weighttongyi-qianwen
Description
Qwen2-72B-Instruct is an instruction-tuned language model with 72 billion parameters, supporting a context length of up to 131,072 tokens. It's part of the new Qwen2 series, which has surpassed most open-source models and demonstrates competitiveness against proprietary models across various benchmarks.
Release Date
2024-06-07
Parameters
72.0B
Context Length
33K
Modalities
text
Capability Radar
70
general
80
coding
70
reasoning
34
scienceest.
73
agents
0
multimodal
Science is estimated from LLM Stats science scores or reasoning when no dedicated science benchmarks are available.
Rankings
No ranking data available
Benchmark Scores (LLM Stats)
(LLM Stats (zeroeval))General
C-Eval
83.8%SR
MMLU
82.3%SR
MultiPL-E
69.2%SR
TruthfulQA
54.8%SR
Language
CMMLU
90.1%SR
MMLU-Pro
64.4%SR
Math
GSM8k
91.1%SR
MATH
59.7%SR
TheoremQA
44.4%SR
Reasoning
HellaSwagAI2 (2019)
87.6%SR
HumanEvalOpenAI (2021)
86.0%SR
Winogrande
85.1%SR
BBH
82.4%SR
MBPP
0.80 / 100SR
EvalPlus
0.79 / 100SR
ARC-C
68.9%SR
GPQANYU + Cohere + Anthropic (2023)
42.4%SR
AA Evaluation Indices
(Artificial Analysis)No AA evaluation data available
LLM Stats Category Scores
(LLM Stats (zeroeval))Language80
Code80
Legal70
Math70
Reasoning70
General70
Healthcare70
Finance60
Physics40
Biology40
Chemistry40
Pricing
Input Price$0.36 / 1M tokens
Output Price$0.4 / 1M tokens
Blended Price (3:1)$0.37 / 1M tokens
Speed
No speed data available
Provider Price Ranking
Provider Price Ranking
1 providers
ProviderInputOutput
1AlibabaPRIMARY
$0.36
$0.4
Compare pricing across different API providers for this model.