Model Comparison

Comprehensive side-by-side analysis of model capabilities and performance

Qwen2.5 72B Instruct

Alibaba Cloud / Qwen Team

Qwen2.5 72B Instruct is a language model developed by Alibaba Cloud / Qwen Team. It achieves strong performance with an average score of 77.4% across 14 benchmarks. It excels particularly in GSM8k (95.8%), MT-Bench (93.5%), MBPP (88.2%). It supports a 139K token context window for handling large documents. The model is available through 4 API providers. It's licensed for commercial use, making it suitable for enterprise applications. Released in 2024, it represents Alibaba Cloud / Qwen Team's latest advancement in AI technology.

Qwen3-235B-A22B-Thinking-2507

Alibaba Cloud / Qwen Team

Qwen3-235B-A22B-Thinking-2507 is a language model developed by Alibaba Cloud / Qwen Team. It achieves strong performance with an average score of 69.2% across 25 benchmarks. It excels particularly in MMLU-Redux (93.8%), AIME25 (92.3%), WritingBench (88.3%). It supports a 387K token context window for handling large documents. The model is available through 1 API provider. It's licensed for commercial use, making it suitable for enterprise applications. Released in 2025, it represents Alibaba Cloud / Qwen Team's latest advancement in AI technology.

Qwen2.5 72B Instruct

Alibaba Cloud / Qwen Team

2024-09-19

Qwen3-235B-A22B-Thinking-2507

Alibaba Cloud / Qwen Team

2025-07-25

10 months newer

Pricing Comparison

Cost per million tokens (USD)

Qwen2.5 72B Instruct

$2.55 cheaper

Input:$0.35

Output:$0.40

Qwen3-235B-A22B-Thinking-2507

Input:$0.30

Output:$3.00

Performance Metrics

Context window and performance specifications

Qwen2.5 72B Instruct

Max Context:139.3K

Parameters:72.7B

Qwen3-235B-A22B-Thinking-2507

Larger context

Max Context:387.1K

Parameters:235.0B

Average performance across 4 common benchmarks

Qwen2.5 72B Instruct

Average Score:72.7%

Qwen3-235B-A22B-Thinking-2507

+14.0%

Average Score:86.8%

Performance comparison across key benchmark categories

Qwen2.5 72B Instruct

factuality

86.8%

roleplay

+15.1%

93.5%

math

+29.4%

89.5%

code

+24.0%

77.9%

general

67.0%

Qwen3-235B-A22B-Thinking-2507

factuality

+7.0%

93.8%

roleplay

78.4%

math

60.1%

code

53.9%

general

+10.3%

77.4%

Benchmark Scores - Detailed View

Side-by-side comparison of all benchmark scores

Provider Availability & Performance

Available providers and their performance metrics

Qwen2.5 72B Instruct

4 providers

Together

Throughput: 47 tok/s

Latency: 0.5ms

Hyperbolic

Throughput: 100 tok/s

Latency: 0.5ms

DeepInfra

Throughput: 10 tok/s

Latency: 0.5ms

Fireworks

Throughput: 59 tok/s

Latency: 0.37ms

Qwen3-235B-A22B-Thinking-2507

1 providers

Novita

Qwen2.5 72B Instruct

Avg Score:72.7%

Providers:4

Qwen3-235B-A22B-Thinking-2507

+14.0%

Avg Score:86.8%

Providers:1