Qwen

Qwen2.5 72B Instruct

by Qwentext->text2 endpoints
Neura Intelligence Index
58.0/ 100
Rank
#138
Confidence
high
Benchmarks
12(31% coverage)

Overview

Qwen2.5 72B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2: - Significantly more knowledge and has greatly improved capabilities in coding and...

Capabilities

Text generation
Tool use / Function calling
Structured output
Top-P sampling
Stop sequences
Deterministic seed

Modalities

Input
Text
Output
Text

Technical Specifications

Context Window
32.8K tokens
Max Output
16.4K tokens
Knowledge Cutoff
2024-06-30
Tokenizer
Qwen
Uptime (24h)
9999.5%
Parameters
72B
Blended Cost
$0.9/1M tokens
License
apache-2.0Open Weights
Country
🇨🇳CN
Model Family
Qwen 2.5
Release Date
September 19, 20241y

Supported Parameters

frequency_penaltylogit_biasmax_tokensmin_ppresence_penaltyrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_p

Benchmark Performance

general

BenchmarkScoreSource
Chatbot Arena Elo
1261
aggregated
IFEval
84.2
aggregated
MMLU-Pro
71.1
aggregated

coding

BenchmarkScoreSource
HumanEval
86.6
aggregated

math

BenchmarkScoreSource
GSM8K
95.8
aggregated
MATH
83.1
aggregated

reasoning

BenchmarkScoreSource
ARC-Challenge
95.3
aggregated
BigBench-Hard
83.5
aggregated
GPQA Diamond
49
aggregated

language

BenchmarkScoreSource
HellaSwag
94
aggregated
WinoGrande
84.8
aggregated

safety

BenchmarkScoreSource
TruthfulQA
62.5
aggregated

Performance Over Time

Pricing

Per 1M tokens
Input$0.36
Output$0.40
Added
September 19, 2024
Last synced 8/27/2026